Introduction

The ocean covers approximately 71% of the earth’s surface and exploration is still in the very beginning stages. The use of HOV for ocean exploration gained prominence in the late 19th century. The development of HOV raises the profile of ocean exploration by allowing humans to participate in expeditions. The ascent and descent mechanisms of the vehicle are crucial factors in HOV. The main ballast system (MBS) is primarily responsible for HOV descent. HOV descent uses a MBS along with wide variety of systems such as variable ballast tank drain off, drop weights, and jettisoning mechanisms1,2. Drop weight mechanisms used in various HOVs include electromagnet, hydraulic, shape memory alloys based and mechanical weight release systems.

The hydraulic drop weight system requires an electric motor, a tank, directional control valves, and a hydraulic cylinder. The hydraulic cylinder is needed to be actuated by a solenoid-actuated direction control valve powered by an electric driven hydraulic pump in order to discard the drop weights. Mechanical drop weight release systems use an electric motor system to operate the mechanism and drop the weights3. Hydraulic and mechanical drop weight release mechanisms consume more power than electromagnet systems. In HOV, power consumption has to be kept as low as possible because the power source is primarily dependent on battery power supply. The current study sheds light on developing a drop weight mechanism that uses less power as possible by incorporating an electromagnet system and a permanent magnet together.

The electromagnet system was widely used in HOV for drop weights and buoyancy adjustments. The steel shots in Nautile HOV were released using an electromagnet system4. The ballast weight in Deep Sea Challenger was released by de-energizing the electromagnet coil5. The electromagnet system was used to change the descent and buoyancy of Triton/limiting factor HOV6. The emergency ballasts weight releasing mechanism in HOV such as MIR I and MIR II, Shinkai 6500, and Jiaolong uses an electromagnet system7,8,9,10,11,12,13. When compared to other drop weight mechanisms, the electromagnet systems compatibility and lower power consumption are significant advantages. However, electromagnet design must be extremely accurate in order to lift and release the required amount of weight and achieve the required buoyancy for descent initiation.

From the literature survey, the design of electromagnet system for the subsea applications was extremely limited. The magnetic field distribution in sea water differs from that of other mediums. Since sea water contains salt, it is a diamagnetic material, causing the magnetic field strength to decrease14. As a result, it is necessary to design an electromagnet system capable of releasing a weight of 25 kg in subsea applications. To improve design and efficiency of the EMDW system, FEA was performed using Infolytica Magnet software. MFD developed during the experiment was measured using a gauss meter and compared to FEA results. In this study, the desirability approach with Response surface methodology (RSM) was used as a tool to design an efficient and optimize EMDW system capable of meeting the required conditions of low power consumption. RSM provides a virtual combination of mathematical and statistical techniques to analyse a specific area of interest or results. For modeling in RSM, the second order polynomial equation was preferred since it uses both curvilinear and non-curvilinear approaches. Second-order polynomial equations can also be used to generate various surfaces and contour plots15,16,17,18. Central composite design was widely used to develop experimental designs for second order polynomial equations, and it was first introduced by Box and Wilson (1951). Optimization was performed to determine the desired output from the combination of input process parameters18. The desirability approach is one of the most commonly employed optimization techniques. The overall value desirability scale runs from 0 to 1. The optimal solution has a value close to one19. In this study, the input process parameters of current, number of turns (NOT), and coil length were varied to determine the optimal process parameter using the RSM and desirability approach. A 50 bar hyperbaric pressure test was conducted to quantify the developed system for 500 m depth real-time subsea application20,21,22.

Response surface methodology

Response surface methodology (RSM) is an optimization tool that uses statistical and mathematical methodologies to develop and enhance processes. The process response is compared and contrasted using residuals, as well as illustration using RSM. Polynomial equations are used in RSM to effectively represent both curvilinear and non-curvilinear models. RSM employs second-order polynomial equations since it has considerable hands-on expertise in real experiments. Equation (1) shows the second order polynomial equation of RSM, where Y- output parameter and X1, X2, X3, X4 – input parameters

$$\rm {Y=\beta_0+\beta_1X_1+\beta_2X_2+\beta_3X_1X_2+\cdots\cdots+Error}$$
(1)

The desired output of the process is determined using optimization techniques. In this study, optimization was carried out using the desirability approach since it is beneficial to compare one or more attributes simultaneously. Three types of optimizations can be done using the desirability approach. (1) Larger is better, (2) Smaller is better, and (3) The optimal value is better. In this study, the optimal value is required for the design that falls between the lower and upper limits of values. The desirable function’s output will range from 1 to 10 using weight functions (1 is the minimum value and 10 is the maximum value). Equation (2) is used for carrying out the desirability approach in this study18,19[,23.

(2)

Human occupied vehicle

In HOV, the power supply for the entire operation is solely reliant on batteries. As a result, the various HOV systems must be able to run on a limited power supply. When the HOV is detached from the mothership, the vehicle begins to float due to the positive buoyancy portrayed in Fig. 1 (a). The primary function of the main ballast system in HOV was to initiate descent. When the main ballast tanks are filled with water, the descent of HOV begins due to a change in HOV buoyancy, as indicated in Fig. 1(b). Following the ocean explorations and sample collections, HOV needs to return to the sea surface by dropping the weight. HOV has used a variety of drop weight mechanisms, including electromagnets, hydraulics, shape memory alloys, and mechanical weight release systems. Given that human life support uses the majority of the battery power, the other systems in HOV should use less power24,25,26. Hence in this study, a new drop weight mechanism is developed that required power supply only at the time of dropping, as shown in Fig. 1(c). Once the drop weight is dropped, the weight of HOV gains positive buoyancy ascent begins, and vehicles return back to the surface as shown in Fig. 1 (d). Figure 2shows the position of the EMDW system in the HOV system27.

Fig. 1
figure 1

(a). HOV after launching from mothership (b) Descent initiation started (c) drop weight released from HOV (d) HOV returns back to surface (NIOT has Autodesk inventor license purchased from Autodesk).

Full size image
Fig. 2
figure 2

Position of EMDW system in HOV (NIOT has Autodesk inventor license purchased from Autodesk).

Full size image

Materials and methods

The various components and cross sectional view of EMDW system is shown in Fig. 3(a) and (b). The core material is a crucial component in an EMDW system due to the fact that it should be capable of producing the necessary MFD for the release of the drop weight (25 kg). The copper coil will be coiled around the core material. The objective of this study was to develop an EMDW system that has capability to develop the required MFD with less power. The core material should have a higher ferrous content or permeability in order to produce increased MFD with less power. The materials permeability will be dependent on a number of factors, including its microstructure, crystal structure, and chemical composition28,15. A pilot experiment was conducted to evaluate the increased MFD produced by two different core materials, mild steel and EN 1 A alloy, with the same amount of power supply. To carry out pilot experiments, a provision has been made in the top flange of the EMDW system as shown in Fig. 3 (a) component no. 3 with a threaded portion to connect the core materials. For the pilot investigation, a 50-turn copper coil with SWG 16 was employed, with a current of 10 A and a voltage of 24 V DC given to both core materials. MFD is measured using a Gauss meter. Table 1indicates the findings of the pilot study. With the same power and number of coil turns are used, EN 1 A alloy produces a greater MFD of 0.286 T than mild steel core material with 0.049 T. This experimental investigation clearly demonstrates that EN 1 A alloy renders higher MFD due to its increased permeability28,15. As a result, EN 1 A alloy was chosen as the core material in this study to develop the EMDW system.

Fig. 3
figure 3

(a) Components of EMDW system (b) Cross sectional view of EMDW system.

Full size image
Table 1 Magnetic flux density.
Full size table

Working of EMDW system

The traditional drop weight release mechanism, which employs an electromagnet require the use of a continuous power supply. As a result, the use of electromagnet in HOS is limited. A new idea for dropping the weight is proposed and implemented by an augmented pair of electromagnet and permanent magnet as shown in Fig. 4 (a)-(c). A weight of 25 kg is lifted initially by incorporating a permanent magnet between the bottom surface of the core material and the top surface of the drop weight as shown in Fig. 4 (a).The permanent magnet used in this study was a neodymium type magnet with an MFD of 0.4 T. When power is supplied, the electromagnet will begin to be energized in such a way that the MFD developed is sufficient to drop weights as shown in Fig. 4 (b). The weight of 25 Kg is dropped from the EMDW system due to the repulsive action between the permanent magnet and electromagnet, as shown in Fig. 4 (c). Like poles repel each other, causing the electromagnet and permanent magnet to act repulsively. In contrast to the continuous power requirement of commercially available electromagnets, the EMDW system only needs the power supply at the moment of cutoff. In this study, a small drawback is that the permanent magnet also gets dropped alongside with 25 kg drop weight.

Fig. 4
figure 4

(a) EMDW System, (b) power supply was given to the electromagnet coil and (c) Dropping of 25 kg weight form EMDW.

Full size image

Results and discussion

Numerical calculation

The amount of mFD required to be developed for dropping the weight of 25 kg has been calculated by the Eq. (3). The gravitational force developed due to drop weight of 25 kg is calculated using Eq. (4)29,30,31. The maximum gravitational force generated by drop weight is 242.25 N (25 kg). The permanent magnet develops an adhesive force of 490 N. Therefore, MFD generated by EMDW system should be the greater than the difference between gravitational force of the load and adhesion force produced by the permanent magnet i.e., 244.75 N (490–242.25 N). Substituting the pull force value needed to produce by EMDW system in Eq. (3), MFD to be developed for dropping a weight of 25 kg is 0.078 Tesla.

$$\rm {F=\frac{1}{2}\times \frac{{{B}^{2}}}{\mathop{\mu }_{0}}\times A}$$
(3)
$$\rm {F=m \times g}$$
(4)

Where, F- Weight required to drop (245.25 N (25 kg), B- Magnetic Flux Density (Tesla), µo-Magnetic Permeability in Space (4∏ × 10−7), A- Surface Area (m2), m is the mass of the load (25 kg), g-acceleration due to gravity (9.81 m/Sect. 2).

Parametric optimization of electromagnet system

In this study, MFD produced for different combinations of input process parameters were determined using FEA. The input design parameters that were taken into account for the optimization were the magnet core length, number of turns (NOT), and current. The magnetization B-H curve of the EN 1 A alloy was used as the input for the FEA in the Infolytica magnet software22. For FEA, the air medium has been used in every component of the EMDW system. Initially, MFD is calculated through numerical calculations, the optimized process parameters must produce an MFD of 0.078 T as the minimum requirement for dropping a weigh of 25 kg using EMDW system. The central composite design was used to develop the experiment level, as shown in Table 2. Three levels of input process parameters were taken in to considerations for parametric optimization. Table 3 displays the corresponding experimental design that had been developed using Design Expert software. Response surface optimization using the desirability approach was employed in this study to determine the optimized values for the design of EMDW system.

Table 2 Level of experiments.
Full size table
Table 3 Design of experiments.
Full size table

Plots analysis on MFD

The mean effect plot of current clearly shows that MFD begins to rise as current increases from 10 to 20 A and then increases dramatically at 30 A, as shown in Fig. 5 (a). However for NOT, MFD increases linearly from 10 to 30 A as shown in Fig. 5 (b). As shown in Fig. 5 (c), MFD begins to increase up to a coil length of 100 mm before decreasing to 110 mm. Mean effect plot of MFD shows the clear path way for the designing of EMDW system needed to be carried out.

Fig. 5
figure 5

Mean Effect plot for various process parameters (a) Current Vs MFD, (b) NOT Vs MFD, and (c) Coil Length Vs MFD.

Full size image

Analysis of Variance (ANOVA) for MFD is given in Table 4, where an F-value of 676.48 is noted for the model of EMDW system. For such a large F-value, there is a 0.01% chance it will occur due to noise. With a P-value less than 0.05, it indicates the models are significant factors with A (current), B (NOT), C (coil length), AB, BC, CA, A2, B2, and C2were found to be significant. R-squared, adjusted R-squared, and adequate precision values were 99.92, 99.97, and 103.299 respectively which were found to be considerably good. The signal-to-noise ratio can be measured by using adequate precision; usually, greater than 4 is desirable. The value of 103.299 indicates the model can be sufficient enough to be used for the design space18,19.

Table 4 ANOVA table.
Full size table

Adequacy testing of RSM model

Before proceeding with the optimization, the model needed to be checked for model adequacy to figure out the level of errors in experimentation. The model adequacy tests were carried out using residual values, which are also useful for determining the relationship between input and output process parameters. Figure 6 (a) shows that the actual and predicted values of MFD using RSM were closely matched, indicating a 45º inclined distribution. There were no trends observed because the values were distributed randomly with externally studentized residuals, which holds valid for the experimentation being carried out as shown in Fig. 6 (b). Internally studentized residuals were also appropriate in this experiment, as the values fell within the range shown in Fig. 6 (c). The normality values were along the line, as shown in Fig. 6(d), indicating that the experiment design is clear of any normality problems18,19,32.

Fig. 6
figure 6

Residual plot of MFD (a) Actual Vs predicted values (b) Run Number Vs. Externally studentized residuals (c) Predicted Vs. Internally studentized residuals and (d) Normal vs. Residuals.

Full size image

Using the desirability approach, the required MFD was set to 0.078 T in design expert software, and the corresponding optimized input process parameters were obtained, as shown in Fig. 7. A separate experiment setup was developed using optimized values for input process parameters from the desirability approach, with a coil length of 106 mm, current of 20 A and a NOT of 18. The experiment was conducted to determine the level of deviation between FEA optimized and experimental values. The experimental findings were illustrated in Table 5. The percentage error between the experimental and numerical approaches is approximately 7.32%, which is minor for acceptance. When the current reaches value of 20 A and 18 NOT with coil length of 106 mm, EMDW system has the sufficient enough MFD to drop the weight of 25 kg. Knowing that the permanent magnet is present along with the drop weight, the exo structure of the HOS should be made of non magnetic materials such as titanium alloys to prevent the permanent magnet from sticking in other parts after dropping. When compared to a commercial electromagnet system, the heat generated by the EMDW system was significantly higher. However, mainly because the EMDW system is designed for 500 m water depth, heat will be dissipated along with the water itself, and no additional care/system is required.

Fig. 7
figure 7

Optimization using desirability approach.

Full size image
Table 5 Comparison of MFD with experimental and FEA value.
Full size table

MFD developed at the bottom of the core material (EN1A alloy) as shown in Fig. 8. The above findings indicate that the EMDW system is capable of dropping weights of up to 25 kg due to the repulsive action between the permanent magnet and the electromagnet. The continuous power supply required for electromagnets has been completely eliminated through the development of the EMDW system, and power will be supplied only at the cut off. Due to the fact that the primary source of HOS is batteries, power savings are essential. Therefore, in HOS the EMDW system can be effectively incorporated instead of the electromagnet.

Fig. 8
figure 8

FEA results of EMDW system for optimized process parameters obtained from Response surface optimization.

Full size image

Hyperbaric chamber pressure testing

Hyperbaric chamber test was used to validate the system operation at sea conditions20,21,22. Hyperbaric chamber pressure testing was carried at a pressure of 50 bar, as shown in Fig. 9(a)-(d). The hyperbaric chamber which has an inner diameter of 400 mm and a height of 500 mm was developed in-house by NIOT Chennai to carry out the test at 50 bar fluid pressure. Water was used as the pressurizing medium in the hyperbaric chamber. To increase the pressure in the chamber and hold it for the required time, a manual hand pump was used. As apparent from Fig. 9 (a), the EMDW system was bolted to the structural frame and a weight of 25 kg was attached and held in place with the help of a permanent magnet. To quantify the EMDW system for 500 m, the system must be effectively operating at 50 bar pressure (Fig. 9 (c)). The coil inside EMDW system has been effectively sealed with O-ring seals on both the top and bottom of the flange in order to prevent the ingress of water. EMDW system was compensated with hydraulic pressure compensated system to nullify the effect produced by water pressure. EMDW system is given a current of 20 A and a voltage of 24 V DC to drop the weight of 25 Kg due to pole repulsive action between permanent magnet and electromagnet in normal conditions without water immersion33. The current of 20 A and voltage of 24 V is found to be sufficient for dropping the weight of 25 Kg from the EMDW system and initiate the descent of HOS from the subsea surface.

Fig. 9
figure 9

(a) System functionality test, (b) System working test in water, (c) Working fluid pressure at hyperbaric chamber, and (d) System working test at rated pressure of 50 bar.

Full size image

Conclusion

Following the development of the EMDW system, several significant discoveries were made, as listed below.

  1. 1.

    The pilot experimentation clearly shows that the MFD produced by EN 1 A alloy is higher than mild steel for the same amount of power. EN 1 A alloy produces higher MFD due to its higher permeability.

  2. 2.

    EMDW system operates on the like pole repulsive effect of an electromagnet and a permanent magnet attached together.

  3. 3.

    The design and development of the EMDW system eliminates the need for continuous power supply for the electromagnet, requiring it only when the weight is dropped.

  4. 4.

    FEA modelling shows that current and NOT were key design variables for the EMDW system.

  5. 5.

    The desirable approach of RSM findings indicate that the EMDW system requires 20 A, 18 NOT, 106 mm coil length, and 24 V DC to drop a weight of 25 kg.

  6. 6.

    A hyperbaric chamber test at 50 bar fluid pressure validated the EMDW system for a rated depth of 500 m. The EMDW system was proven to be successful at lowering 25 kg of weight using a power source of 24 V DC and 20 A.