Method and device for evaluating application effect of ship energy efficiency measures

Abstract

The application provides a kind of ship energy efficiency measure application effect evaluation method and device, evaluation method includes: obtaining ship multi-source heterogeneous data;Multi-source heterogeneous data is fused and is handled, and target data is obtained;Target data is divided into first data and second data;First data is input into second oil consumption prediction model and is carried out oil consumption prediction, and first prediction result is obtained;Second data is input into first oil consumption prediction model and is carried out oil consumption prediction, and second prediction result is obtained;First oil consumption prediction model and second oil consumption prediction model are obtained by training preset machine learning model;According to first prediction result and first actual oil consumption value, first evaluation result is obtained;According to second prediction result and second actual oil consumption value, second evaluation result is obtained;According to first evaluation result and second evaluation result, target evaluation result is obtained, and output is carried out.The application can improve the efficiency and convenience of ship energy efficiency measure application effect evaluation, reduce evaluation cost.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer information processing, in particular to a method and device for evaluating the application effect of ship energy efficiency measures. BACKGROUND

[0002] Ship technical energy efficiency measures include hull lines optimization, ship lightweight, air film drag reduction, coating drag reduction, high-efficiency propeller, hydrodynamic energy-saving device, waste heat recovery, etc., and operational energy efficiency measures include weather route planning, speed optimization, optimal shaft power, trim optimization, route optimization, hull cleaning, digital solution, etc. During the operation of a ship, multiple technical and operational energy efficiency measures are usually applied simultaneously. How to accurately evaluate the comprehensive effect of applying one or more energy-saving and carbon-reducing measures in actual operation of a ship has become a key to whether the shipping industry can achieve net zero emissions in the future.

[0003] The existing methods for evaluating the application effect of ship energy efficiency measures mainly include ship energy efficiency dynamic simulation method, ship model test method and real ship trial method. The ship energy efficiency dynamic simulation method requires high computing resources and needs high-performance computers to process complex simulation tasks. The model precision is limited, and simplification assumptions and turbulence modeling may affect the accuracy of the results. It is difficult to verify and confirm, and it requires rich experience and data support to ensure the effectiveness and accuracy of the model. The user skill requirement is high, and professional knowledge and experience are needed to set up and interpret the results. The ship model test method produces scale effects, and although it follows the similarity law, it is difficult to completely eliminate the influence of scale differences in actual operation. The cost is high, and a large amount of funds is needed for high-quality model production, maintenance and operation of large experimental facilities. It takes a long time to complete a series of experiments from model production. The range is limited and cannot fully represent the behavior of full-size ships, especially when non-linear effects are involved. The real ship trial method is costly, involving fuel, manpower, time and potential risk costs. It is weather limited and needs to choose the right weather window. The amount of data is limited, and compared with ship energy efficiency dynamic simulation and ship model test, the amount of data obtained is relatively small. The trial conditions and sea conditions are limited, and it is difficult to cover the actual operating conditions of the ship. The existing evaluation methods are based on simulation and test after simplification of the actual operating conditions of the ship. The application effect of energy efficiency measures calculated by the methods has a certain reference value, but it is difficult to fully consider the complex actual operating conditions of the ship, and different ship types and tonnages need to be re-modeled and calculated, which is costly and time-consuming, and non-professional personnel cannot evaluate it. SUMMARY

[0004] The technical problem to be solved by the present application is to provide a method and device for evaluating the application effect of ship energy efficiency measures. The efficiency and convenience of evaluating the application effect of ship energy efficiency measures can be improved, and the evaluation cost can be reduced.

[0005] To solve the above technical problems, the technical solutions of the present application are as follows:

[0006] A method for evaluating the application effect of ship energy efficiency measures, comprising:

[0007] Obtaining multi-source heterogeneous data of a ship;

[0008] Fusing the multi-source heterogeneous data to obtain target data;

[0009] Dividing the target data into first data before the application of energy efficiency measures and second data after the application of energy efficiency measures according to the time of the application of energy efficiency measures;

[0010] Inputting the first data into a second oil consumption prediction model to perform oil consumption prediction and obtain a first prediction result; the second oil consumption prediction model is obtained by training a preset machine learning model according to a second data set;

[0011] Inputting the second data into a first oil consumption prediction model to perform oil consumption prediction and obtain a second prediction result; the first oil consumption prediction model is obtained by training a preset machine learning model according to a first data set;

[0012] Obtaining a first evaluation result according to the first prediction result and a first actual oil consumption value;

[0013] Obtaining a second evaluation result according to the second prediction result and a second actual oil consumption value;

[0014] Obtaining a target evaluation result according to the first evaluation result and the second evaluation result, and outputting.

[0015] Optionally, the fusion processing of the multi-source heterogeneous data to obtain target data comprises:

[0016] Performing data cleaning on the multi-source heterogeneous data to obtain first intermediate data;

[0017] Performing format conversion on the first intermediate data to obtain second intermediate data;

[0018] Performing data fusion processing on the second intermediate data to obtain target data.

[0019] Optionally, the training process of the first oil consumption prediction model comprises:

[0020] Obtaining a first data set, the first data set comprising a first training set and a first validation set;

[0021] Inputting data in the first training set into an input layer of a preset machine learning model to perform processing and obtain a first output;

[0022] input the first output into a loop processing layer of the preset machine learning model for processing to obtain a second output;

[0023] input the second output into an output layer of the preset machine learning model for processing to obtain a first prediction result;

[0024] optimize the preset machine learning model according to the first prediction result, the first verification set and mean absolute percentage error to obtain a first fuel consumption prediction model.

[0025] Optionally, the training process of the second fuel consumption prediction model comprises:

[0026] obtain a second data set, the second data set comprising a second training set and a second verification set;

[0027] input data in the second training set into an input layer of the preset machine learning model for processing to obtain a third output;

[0028] input the third output into a loop processing layer of the preset machine learning model for processing to obtain a fourth output;

[0029] input the fourth output into an output layer of the preset machine learning model for processing to obtain a second prediction result;

[0030] optimize the preset machine learning model according to the second prediction result, the second verification set and mean absolute percentage error to obtain a second fuel consumption prediction model.

[0031] Optionally, inputting the first data into the second fuel consumption prediction model for fuel consumption prediction to obtain a first prediction result comprises:

[0032] input the first data into an input layer of the second fuel consumption prediction model for processing to obtain a third output;

[0033] input the third output into a loop processing layer of the preset machine learning model for processing to obtain a fourth output;

[0034] input the fourth output into an output layer of the preset machine learning model for processing to obtain a first prediction result.

[0035] Optionally, inputting the second data into the first fuel consumption prediction model for fuel consumption prediction to obtain a second prediction result comprises:

[0036] input the second data into an input layer of the first fuel consumption prediction model for processing to obtain a first output;

[0037] input the first output into a loop processing layer of the preset machine learning model for processing to obtain a second output;

[0038] input the second output into an output layer of the preset machine learning model for processing to obtain a second prediction result.

[0039] Optionally, according to the first prediction result and a first actual fuel consumption value, a first evaluation result is obtained, including:

[0040] According to the first predicted fuel consumption, the first actual fuel consumption and the first mean absolute percentage error, the first evaluation result is obtained.

[0041] According to the second prediction result and a second fuel consumption value, a second evaluation result is obtained, including:

[0042] According to the second predicted fuel consumption, the second actual fuel consumption and the second mean absolute percentage error, the second evaluation result is obtained.

[0043] Optionally, according to the first evaluation result and the second evaluation result, a target evaluation result is obtained, including:

[0044] According to the first evaluation result and the second evaluation result, an intermediate evaluation result is obtained.

[0045] According to the intermediate evaluation result and an index weight, the target evaluation result is obtained.

[0046] Embodiments of the present application also provide a ship energy efficiency measure application effect evaluation device, including:

[0047] An acquisition module is configured to acquire ship multi-source heterogeneous data.

[0048] A processing module is configured to perform fusion processing on the multi-source heterogeneous data to obtain target data; divide the target data into first data before application of an energy efficiency measure and second data after application of the energy efficiency measure according to time of application of the energy efficiency measure; input the first data into a second fuel consumption prediction model to perform fuel consumption prediction and obtain a first prediction result; the second fuel consumption prediction model is obtained by training a preset machine learning model according to a second data set; input the second data into a first fuel consumption prediction model to perform fuel consumption prediction and obtain a second prediction result; the first fuel consumption prediction model is obtained by training the preset machine learning model according to a first data set; obtain a first evaluation result according to the first prediction result and a first actual fuel consumption value; obtain a second evaluation result according to the second prediction result and a second actual fuel consumption value; obtain a target evaluation result according to the first evaluation result and the second evaluation result, and output.

[0049] Embodiments of the present application also provide a computing device readable storage medium, which stores a program, and the program is executed by a processor to implement the ship energy efficiency measure application effect evaluation method.

[0050] The above technical solutions of the present application have at least the following technical effects:

[0051] The above ship energy efficiency measure application effect evaluation method of the present application can improve the efficiency and convenience of ship energy efficiency measure application effect evaluation, and reduce the evaluation cost. BRIEF DESCRIPTION OF DRAWINGS

[0052] Figure 1 is a flowchart of the ship energy efficiency measure application effect evaluation method of the present application;

[0053] Figure 2 is a schematic diagram of the machine learning model of the ship energy efficiency measure application effect evaluation method of the present application;

[0054] Figure 3 is an implementation schematic diagram of the ship energy efficiency measure application effect evaluation method of the present application;

[0055] Figure 4 is a schematic diagram of the ship energy efficiency measure application effect evaluation device of the present application. DETAILED DESCRIPTION

[0056] Exemplary embodiments of the present application will be described in greater detail below with reference to the accompanying drawings. Although exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present application can be more thoroughly understood and the scope of the present application can be accurately conveyed to those skilled in the art.

[0057] As shown in Figure 1 , an embodiment of the present application proposes a ship energy efficiency measure application effect evaluation method, comprising:

[0058] Step S1, acquiring ship multi-source heterogeneous data;

[0059] Step S2, the multi-source heterogeneous data is fused to obtain target data;

[0060] Step S3, the target data is divided into first data before the energy efficiency measure is applied and second data after the energy efficiency measure is applied according to the time when the energy efficiency measure is applied;

[0061] Step S4, the first data is input into a second fuel consumption prediction model for fuel consumption prediction to obtain a first prediction result; the second fuel consumption prediction model is obtained by training a preset machine learning model according to a second data set;

[0062] Step S5, the second data is input into a first fuel consumption prediction model for fuel consumption prediction to obtain a second prediction result; the first fuel consumption prediction model is obtained by training a preset machine learning model according to a first data set;

[0063] Step S6, a first evaluation result is obtained according to the first prediction result and a first actual fuel consumption value;

[0064] Step S7, a second evaluation result is obtained according to the second prediction result and a second actual fuel consumption value;

[0065] Step S8, a target evaluation result is obtained according to the first evaluation result and the second evaluation result, and output.

[0066] In this embodiment, as Figure 1As shown in the ship energy efficiency measure application effect evaluation method, first, the multi-source heterogeneous data of the ship is obtained, and the multi-source data that affects the ship energy efficiency result is selected, which is directly related to the result of the ship energy efficiency; then, the obtained multi-source heterogeneous data is screened, removed, unified, and data fusion is carried out, and a basic data is formed according to the corresponding association relationship, and target data is obtained; thirdly, the target data is divided into first data and second data according to the time of applying the energy efficiency measure; fourthly, the first data is processed by using the second oil consumption prediction model to predict the oil consumption, and the first prediction result is obtained; the second oil consumption prediction model is obtained by training the preset machine learning model according to the second data set; the second data is processed by using the first oil consumption prediction model to predict the oil consumption, and the second prediction result is obtained; the first oil consumption prediction model is obtained by training the preset machine learning model according to the first data set; next, according to the first prediction result and the first actual oil consumption value, the prediction result is compared with the actual result, and the first evaluation result is obtained by combining the error factor; according to the second prediction result and the second actual oil consumption value, the prediction result is compared with the actual result, and the second evaluation result is obtained by combining the error factor; finally, according to the first evaluation result and the second evaluation result, the appropriate evaluation data is selected, the target evaluation result is obtained, and the output is obtained; the scheme of the application can realize the evaluation of the application effect of the energy efficiency measure of the ship under different speeds, different cargo water levels and different sea conditions, and realize the comprehensive effect evaluation of the application of one or more technical energy efficiency and operation energy efficiency measures of the ship; the method has low cost, high efficiency and short time, and can provide a convenient ship energy efficiency measure effect evaluation tool for non-professionals, and save a lot of human resource cost.

[0067] In the embodiment, the obtained multi-source heterogeneous data includes: ship automatic identification system data, ship navigation data, marine meteorological sea condition data and ship static data;

[0068] The ship static data is mainly the basic technical parameters of the ship, such as the main dimension parameters, the main and auxiliary machine parameters, etc., and the main indicators are as follows:

[0069]

[0070] The ship automatic identification system data is the information data of the ship automatic identification system, such as the position, speed and heading of the marine ship, and the main indicators are as follows:

[0071]

[0072] The marine meteorological sea condition data is high spatio-temporal resolution marine meteorological reanalysis grid data, including wind, wave, current and other sea condition data, and the main indicators are as follows:

[0073]

[0074]

[0075] Ship navigation data is various data generated during ship navigation, including ship navigation attitude data, ship main equipment operation data, fuel consumption data, ship body data, and fouling data, and the main indicators are as follows:

[0076]

[0077]

[0078] In an optional embodiment of the application, in step S2, the multi-source heterogeneous data is fused to obtain target data, including:

[0079] In step S21, the multi-source heterogeneous data is cleaned to obtain first intermediate data.

[0080] In step S22, the first intermediate data is converted in format to obtain second intermediate data.

[0081] In step S23, the second intermediate data is fused to obtain target data.

[0082] In this embodiment, the ship-related data is large in quantity and has much noise data, so the multi-source heterogeneous data needs to be preprocessed. First, the multi-source heterogeneous data is cleaned to remove abnormal data, error data and redundant data that are obviously out of the normal range to obtain first intermediate data, such as preprocessing abnormal data that exceeds the speed threshold and exceeds the latitude and longitude range. Then, the first intermediate data is converted in format to unify the data collected by different sensors into a standard format, for example, converting weight data into kg and time data into YYYY-MM-DD HH:MM:SS format to obtain second intermediate data. Finally, the second intermediate data is fused. First, according to time and latitude and longitude information, the ship automatic identification system data and marine weather and sea condition data are matched to concatenate the data according to time and latitude and longitude to obtain intermediate fusion data. Then, according to the ship marine mobile identification code (MMSI) and the International Maritime Organization identification number (IMO), the ship static data is matched for the intermediate fusion data to concatenate the data according to MMSI and IMO information to obtain target data.

[0083] In an optional embodiment of the application, in step S5, the training process of the first fuel consumption prediction model includes:

[0084] In step S511, a first data set is obtained, and the first data set includes a first training set and a first validation set.

[0085] Step S512, inputting the data in the first training set into an input layer of the preset machine learning model for processing to obtain a first output;

[0086] Step S513, inputting the first output into a recurrent processing layer of the preset machine learning model for processing to obtain a second output;

[0087] Step S514, inputting the second output into an output layer of the preset machine learning model for processing to obtain a first prediction result;

[0088] Step S515, optimizing the preset machine learning model according to the first prediction result, the first verification set and a mean absolute percentage error to obtain a first fuel consumption prediction model.

[0089] In this embodiment, first, a first data set is acquired, which is objective data accumulated historically before application of energy efficiency measures; the first data set is divided into a first training set and a first verification set according to a preset ratio, a preset machine learning model is trained with a mean absolute percentage error as a model evaluation index to obtain a first fuel consumption prediction model.

[0090] Specifically, first, based on the first data set dataA1, the data is randomly divided into the first training set and the first verification set according to a ratio of 7:3, a machine learning model is selected as a prediction model, and the model is trained with the first training set and the first verification set; a mean absolute percentage error (MAPE) is selected as a model evaluation index, and an optimal model with a sample point fuel consumption MAPE less than 3% and a segmented voyage and segmented speed aggregated fuel consumption MAPE less than 1% is selected as a first fuel consumption prediction model (modelA) for fuel consumption before application of energy efficiency measures, and the MAPE is selected as a prediction error E of the model modelA modelA，i，j The format is as follows:

[0091]

[0092] Further, as shown in the preset neural network Figure 2 , the first training set data is divided into batches, and each batch of data is input into the preset neural network to obtain the output result of the model. For example, if the training set has 1000 samples and the single batch size is set to 100, then 100 samples are input into the model as a group each time. The input layer of the preset machine learning model receives the training set data in batches and processes to obtain a first output; then the first output is input into the recurrent processing layer of the preset machine learning model for processing to obtain a second output; the recurrent processing layer includes a forget gate rt, an input gate zt, a current neuron state h't, and the forget gate rt is obtained from the following formula:

[0093] rt = sigmoid(Qr·[h t-1 ,xt])

[0094] Where sigmoid is the activation function, Qr is the weight matrix, and h t-1 The state of the previous neuron is given by , and xt is the input value of the current neuron.

[0095] The input gate zt is obtained by the following formula:

[0096] zt=sigmoid(Qz·[h t-1,xt])

[0097] Where sigmoid is the activation function, Qz is the weight matrix, and h t-1 The state of the previous neuron is given by , and xt is the input value of the current neuron.

[0098] The current neuron state h′t is obtained by the following formula:

[0099] h′t=tanh(Q·[rt*h t-1 ,xt])

[0100] Where tanh is the activation function, Q is the weight matrix, rt is the forget gate output value, and h t-1 The state of the previous neuron is given by , and xt is the input value of the current neuron.

[0101] The current neuron's output value ht is obtained by the following formula:

[0102] ht=(1-zt)*h t-1 +zt*h′t])

[0103] Where zt is the input gate, h t-1 h't represents the state of the previous neuron, and h't represents the state of the current neuron.

[0104] By comparing the model output with the corresponding true labels (true values ​​in the validation set) using the selected loss function, the loss value is calculated to measure the quality of the current model's predictions.

[0105] This embodiment uses the mean absolute percentage error as the loss function:

[0106]

[0107] Among them, E modelA，i，j Let y be the prediction error of model A under the conditions of speed i and draft j, n be the number of samples, and y be the prediction error of model A. i，j Let (i, j) be the actual observed value. The predicted value corresponding to the actual observation value under the condition of (i, j).

[0108] According to the calculated loss value, the chain rule is used for back propagation, the gradient corresponding to each parameter is calculated, and then the sample gradient is calculated through a stochastic gradient optimization algorithm, and the sample gradient calculation formula of the weight parameter w is as follows:

[0109] w = w - η·∧ w E(w)

[0110] Wherein, w is a weight parameter, η is a model learning rate, used to control the step size of each parameter update, ∧ w E(w) is the gradient of the loss function E with respect to w.

[0111] According to these gradients, the parameters of the model are updated, so that the model is continuously adjusted in the direction of loss reduction. This process is repeated for a predetermined number of rounds to avoid under-training or overfitting. After each round of training, the performance of the model on the current round is evaluated using the validation set, such as calculating the accuracy, loss value and other indicators on the validation set, so as to adjust the hyperparameters in time or stop training in advance. If the performance of the validation set no longer improves or even decreases, overfitting may occur, and training can be stopped; after training is completed, the second output is input into the output layer of the preset machine learning model for processing to obtain a first fuel consumption prediction model;

[0112] Similarly, other machine learning models are selected, such as decision tree, random forest, K nearest neighbor, elastic network regression, extreme gradient boosting tree, and gradient boosting decision tree, and the training process is similar to the above example, which can also have the same effect.

[0113] In an optional embodiment of the present application, in step S4, the training process of the second fuel consumption prediction model comprises:

[0114] Step S411, obtaining a second data set, the second data set comprising: a second training set and a second validation set;

[0115] Step S412, inputting the data in the second training set into the input layer of the preset machine learning model for processing to obtain a third output;

[0116] Step S413, inputting the third output into the loop processing layer of the preset machine learning model for processing to obtain a fourth output;

[0117] Step S414, inputting the fourth output into the output layer of the preset machine learning model for processing to obtain a second prediction result;

[0118] Step S415, according to the second prediction result, the second verification set and the mean absolute percentage error, the preset machine learning model is optimized to obtain a second fuel consumption prediction model.

[0119] In this embodiment, similar to the first fuel consumption prediction model training process, based on the second data set dataB1, the data is randomly divided into a second training set and a second verification set in a ratio of 7:3, a machine learning model is selected as a prediction model, and the model is trained with the second training set and the second verification set; the mean absolute percentage error (MAPE) is taken as the model evaluation index, the optimal model with the sample point fuel consumption MAPE less than 3% and the segmented section and segmented speed summarized fuel consumption MAPE less than 1% is selected as the second fuel consumption prediction model (modelB) before the application of the energy efficiency measure, and the MAPE is taken as the prediction error E of the model modelB modelB，i，j , which is as follows:

[0120]

[0121] Further, the preset neural network is as shown in the following table: Figure 2 The second training set data is divided into batches, and each batch of data is input into the preset neural network to obtain the output result of the model. For example, if the training set has 1000 samples and the single batch size is set to 100, then 100 samples are input into the model as a group each time. The input layer of the preset machine learning model receives the training set data in batches and processes to obtain a third output; then the third output is input into the preset machine learning model for processing to obtain a fourth output; the loop processing layer includes a forgetting gate rt, an input gate zt, a current neuron state h't, and the forgetting gate rt is obtained by the following formula:

[0122] rt=sigmoid(Qr·[h t-1 ,xt])

[0123] Wherein, sigmoid is an activation function, Qr is a weight matrix, h t-1 is the previous neuron state, and xt is the input value of the current neuron.

[0124] The input gate zt is obtained by the following formula:

[0125] zt=sigmoid(Qz·[h t-1,xt])

[0126] Wherein, sigmoid is an activation function, Qz is a weight matrix, h t-1 is the previous neuron state, and xt is the input value of the current neuron.

[0127] The current neuron state h't is obtained by the following formula:

[0128] h't = tanh(Q · [rt*h t-1 , xt])

[0129] Wherein, tanh is an activation function, Q is a weight matrix, rt is a forgetting gate output value, h t-1 is the last neuron state, and xt is the input value of the current neuron.

[0130] The output value ht of the current neuron is obtained by the following formula:

[0131] ht = (1-zt)*h t-1 +zt*h't])

[0132] Wherein, zt is an input gate, h t-1 is the last neuron state, and h't is the current neuron state.

[0133] By selecting a loss function, the loss value is calculated by comparing the model output result and the corresponding true label (the true value in the validation set), so as to measure the goodness of the current model prediction.

[0134] The mean absolute percentage error is used as the loss function in this embodiment:

[0135]

[0136] Wherein, E modelA，i，j is the prediction error of the prediction model modelA under the condition of speed i and draft j, n is the sample number, y i，j is the actual observation value under the condition of (i, j), is the prediction value corresponding to the actual observation value under the condition of (i, j).

[0137] According to the calculated loss value, the chain rule is used for back propagation to calculate the gradient corresponding to each parameter, and then the sample gradient is calculated by using the stochastic gradient optimization algorithm. The sample gradient calculation formula of the weight parameter w is as follows:

[0138] w = w - η · ∧ w E(w)

[0139] Wherein, w is the weight parameter, η is the model learning rate, which is used to control the step size of each parameter update, ∧ w E(w) is the gradient of the loss function E with respect to w.

[0140] According to the gradient, the parameters of the model are updated, and the model is continuously adjusted in the direction of loss reduction. This process is repeated for a preset number of rounds to avoid under-training or overfitting. After each round of training, the performance of the model on the validation set is evaluated, such as calculating the accuracy, loss value, and other indicators on the validation set, to adjust the hyperparameters in time or stop training in advance. If the performance of the validation set no longer improves or even decreases, overfitting may occur, and training can be stopped. After training is completed, the fourth output is input into the output layer of the preset machine learning model for processing to obtain a second fuel consumption prediction model;

[0141] Similarly, other machine learning models are selected, such as decision tree, random forest, K-nearest neighbor, elastic network regression, extreme gradient boosting tree, and gradient boosting decision tree. The training process is similar to the above example and can achieve the same effect.

[0142] In an optional embodiment of the present application, in step S4, the first data is input into the second fuel consumption prediction model for fuel consumption prediction to obtain a first prediction result, including:

[0143] In step S421, the first data is input into the input layer of the second fuel consumption prediction model for processing to obtain a third output.

[0144] In step S422, the third output is input into the recurrent processing layer of the preset machine learning model for processing to obtain a fourth output.

[0145] In step S423, the fourth output is input into the output layer of the preset machine learning model for processing to obtain a first prediction result.

[0146] In this embodiment, as Figure 3As shown, the target data includes ship data before and after the application of energy efficiency measures, and according to whether the energy efficiency measures are applied to the ship, the data in the target data before the application of energy efficiency measures is taken as first data (data A), and the data in the target data after the application of energy efficiency measures is taken as second data (data B); the first data is divided into small batches (batch), and each batch of data is input into the input layer of the second fuel consumption prediction model for processing to obtain a third output. For example, if the first data has 1000 samples, and the size of a single batch is set to 100, then 100 samples are input into the model as a group each time. The input layer of the second fuel consumption prediction model receives the training set data in batches and processes to obtain a third output; then the third output is input into the cycle processing layer of the second fuel consumption prediction model for processing, and the cycle processing layer includes a forget gate rt, an input gate zt, a current neuron state h't, and a fourth output is obtained; and then the fourth output is input into the output layer of the second fuel consumption prediction model for processing to obtain a first prediction result.

[0147] In an optional embodiment of the present application, in step S5, the second data is input into the first fuel consumption prediction model for fuel consumption prediction to obtain a second prediction result, including:

[0148] In step S521, the second data is input into the input layer of the first fuel consumption prediction model for processing to obtain a first output.

[0149] In step S522, the first output is input into the cycle processing layer of the preset machine learning model for processing to obtain a second output.

[0150] In step S523, the second output is input into the output layer of the preset machine learning model for processing to obtain a second prediction result.

[0151] In this embodiment, as shown in FIG. 6, the second data is input into the input layer of the first fuel consumption prediction model for processing to obtain a first output. Figure 3As shown, the target data includes ship data before and after the application of energy efficiency measures, and according to whether the energy efficiency measures are applied to the ship, the data before the application of energy efficiency measures in the target data is taken as first data (dataA), and the data after the application of energy efficiency measures in the target data is taken as second data (dataB); the second data is divided into batches, and each batch of data is input into the input layer of the first fuel consumption prediction model for processing to obtain a first output. For example, if the second data has 1000 samples, and the size of a single batch is set to 100, then 100 samples are input into the model as a group each time. The input layer of the first fuel consumption prediction model receives the training set data in batches and processes to obtain a first output; then the first output is input into the cycle processing layer of the first fuel consumption prediction model for processing, and the cycle processing layer includes a forget gate rt, an input gate zt, a current neuron state h't, to obtain a second output; and then the second output is input into the output layer of the first fuel consumption prediction model for processing to obtain a second prediction result.

[0152] In an optional embodiment of the present application, in step S6, a first evaluation result is obtained according to the first prediction result and the first actual fuel consumption value, including:

[0153] In step S61, a first evaluation result is obtained according to the first predicted fuel consumption, the first actual fuel consumption and the first mean absolute percentage error;

[0154] In step S7, a second evaluation result is obtained according to the second prediction result and the second fuel consumption value, including:

[0155] In step S71, a second evaluation result is obtained according to the second predicted fuel consumption, the second actual fuel consumption and the second mean absolute percentage error.

[0156] In the embodiment, as shown, Figure 3 first, the first predicted fuel consumption after the application of energy efficiency measures is compared and analyzed with the first actual fuel consumption before the application of energy efficiency measures in dataA to obtain a first evaluation result, and the first evaluation result calculation formula of the application of energy efficiency measures is as follows:

[0157] f modelB，dataA，i，j = |FCP modelB，dataA，i，j -FCR dataA，i，j | / FCR dataA，i，j * 100% - E modelB，i，j

[0158] Wherein, f modelB，dataA，i，j is the energy saving effect (%) of the ship applying energy efficiency measures under the condition of speed i and draft j in data dataA measured by the second model modelB, FCP modelB，dataA，i，jFCR is the total predicted fuel consumption of the second model modelB under the condition of the speed i and the draft j in the first data dataA dataA，i，j E is the total actual fuel consumption under the condition of the speed i and the draft j in the first data dataA modelB，i，j is the prediction error of the second model modelB under the condition of the speed i and the draft j.

[0159] Then, based on the second data dataB, the second predicted fuel consumption before the application of the energy efficiency measure is obtained through the first model modelA, and the second actual fuel consumption after the application of the energy efficiency measure of the second data dataB is compared and analyzed, and the second evaluation result of the application of the energy efficiency measure is calculated according to the following formula:

[0160] f modelA，dataB，i，j =|FCR dataB，i，j -FCP modelA，dataB，i，j | / FCP modelA，dataB，i，j *100%-E modelA，i，j

[0161] Wherein, f modelA，dataB，i，j is the energy saving effect (%) of the ship applying the energy efficiency measure under the condition of the speed i and the draft j in the data dataB, FCR dataB，i，j is the total predicted fuel consumption of the first model modelA under the condition of the speed i and the draft j in the data dataB, FCP modelA，dataB，i，j is the total actual fuel consumption under the condition of the speed i and the draft j in the data dataB, E modelA，i，j is the prediction error of the prediction model modelA under the condition of the speed i and the draft j.

[0162] In an optional embodiment of the present application, in step S8, the target evaluation result is obtained according to the first evaluation result and the second evaluation result, comprising:

[0163] In step S81, the intermediate evaluation result is obtained according to the first evaluation result and the second evaluation result.

[0164] In step S82, the target evaluation result is obtained according to the intermediate evaluation result and the index weight.

[0165] In this embodiment, first, the first evaluation result and the second evaluation result are compared, and the smaller one of the first evaluation result and the second evaluation result is taken as the intermediate evaluation result, and the final effect f i，j of the application of the energy efficiency measure under the condition of the speed i and the draft j is calculated according to the following formula:

[0166] f i，j =min(f modelA，dataB，i，j ,f modelB，dataA，i，j )

[0167] Then, the proportion of the number of turns of the ship of the specific ship type of the shipping industry under the condition of the sailing speed i and the draft j is taken as the weight ω i，j The table can be given as follows:

[0168]

[0169] The weighted average of the application of the energy efficiency measure is calculated as the energy efficiency measure application target evaluation result f, and the calculation formula of the target evaluation result f is:

[0170] f = ∑∑f i，j × ω i，j

[0171] The specific implementation process of the above-mentioned method of the present application is described below:

[0172] Step 111, acquiring multi-source heterogeneous data of the ship;

[0173] Step 112, data cleaning is performed on the multi-source heterogeneous data to obtain first intermediate data;

[0174] Step 113, format conversion is performed on the first intermediate data to obtain second intermediate data;

[0175] Step 114, data fusion processing is performed on the second intermediate data to obtain target data;

[0176] Step 115, the target data is divided into first data before the application of the energy efficiency measure and second data after the application of the energy efficiency measure according to the time of the application of the energy efficiency measure;

[0177] Step 116, acquiring a first data set, the first data set comprising: a first training set and a first validation set;

[0178] Step 117, inputting the data in the first training set into the input layer of the preset machine learning model for processing to obtain a first output;

[0179] Step 118, inputting the first output into the loop processing layer of the preset machine learning model for processing to obtain a second output;

[0180] Step 119, inputting the second output into the output layer of the preset machine learning model for processing to obtain a first prediction result;

[0181] Step 120, according to the first prediction result, the first validation set and the mean absolute percentage error, the preset machine learning model is tuned to obtain a first fuel consumption prediction model;

[0182] Step 121, acquiring a second data set, the second data set comprising: a second training set and a second validation set;

[0183] Step 122, inputting the data in the second training set into the input layer of the preset machine learning model for processing to obtain a third output;

[0184] Step 123, inputting the third output into the loop processing layer of the preset machine learning model for processing to obtain a fourth output;

[0185] Step 124, inputting the fourth output into the output layer of the preset machine learning model for processing to obtain a second prediction result;

[0186] Step 125, according to the second prediction result, the second verification set and the mean absolute percentage error, the preset machine learning model is optimized to obtain a second fuel consumption prediction model;

[0187] Step 126, inputting the first data into the input layer of the second fuel consumption prediction model for processing to obtain a third output;

[0188] Step 127, inputting the third output into the loop processing layer of the preset machine learning model for processing to obtain a fourth output;

[0189] Step 128, inputting the fourth output into the output layer of the preset machine learning model for processing to obtain a first prediction result;

[0190] Step 129, inputting the second data into the input layer of the first fuel consumption prediction model for processing to obtain a first output;

[0191] Step 130, inputting the first output into the loop processing layer of the preset machine learning model for processing to obtain a second output;

[0192] Step 131, inputting the second output into the output layer of the preset machine learning model for processing to obtain a second prediction result;

[0193] Step 132, according to the first predicted fuel consumption, the first actual fuel consumption and the first mean absolute percentage error, a first evaluation result is obtained;

[0194] Step 133, according to the second prediction result and the second fuel consumption value, a second evaluation result is obtained, including:

[0195] Step 134, according to the second predicted fuel consumption, the second actual fuel consumption and the second mean absolute percentage error, a second evaluation result is obtained;

[0196] Step 135, according to the first evaluation result and the second evaluation result, an intermediate evaluation result is obtained;

[0197] In step 136, a target evaluation result is obtained according to the intermediate evaluation result and the index weight.

[0198] The scheme of the present application comprehensively utilizes multi-source heterogeneous data before and after the application of energy efficiency measures by the operating ship, fuses ship AIS data, ship navigation data, marine weather and sea condition data and ship static data, and constructs a ship energy efficiency measure application effect evaluation method based on a data model. By constructing two high-precision oil consumption prediction models, the evaluation effect deviation caused by the navigation condition difference of the baseline group and the control group ship is eliminated by applying the exchange data, the influence of more actual ship navigation complex working conditions on the energy efficiency technology application effect evaluation is fully considered, and a calculation method of the overall effect of the energy efficiency measure is proposed.

[0199] The scheme of the present application establishes an oil consumption prediction model before and after the application of energy efficiency measures by the ship based on the actual ship navigation data, fully considers the complex working conditions of the actual ship navigation, and establishes the oil consumption prediction model by the method of machine learning; the application effect of the energy efficiency measure of the ship under different navigation speeds, different drafts (cargo) and different sea condition conditions can be evaluated; the comprehensive effect evaluation of one or more technical energy efficiency and operating energy efficiency measures of the ship can be realized; the method of the present application has low cost, high efficiency and short time consumption, and can provide a convenient ship energy efficiency measure effect evaluation tool, and save a large amount of human resource cost.

[0200] As shown in Figure 4 The embodiment of the present application also provides a ship energy efficiency measure application effect evaluation device 30, which comprises:

[0201] The acquisition module 41 is used for acquiring multi-source heterogeneous data of the ship.

[0202] The processing module 42 is used for fusing and processing the multi-source heterogeneous data to obtain target data; the target data is divided into first data before the application of the energy efficiency measure and second data after the application of the energy efficiency measure according to the time of the application of the energy efficiency measure; the first data is input into a second oil consumption prediction model for oil consumption prediction to obtain a first prediction result; the second oil consumption prediction model is obtained by training a preset machine learning model according to a second data set; the second data is input into a first oil consumption prediction model for oil consumption prediction to obtain a second prediction result; the first oil consumption prediction model is obtained by training the preset machine learning model according to a first data set; a first evaluation result is obtained according to the first prediction result and a first actual oil consumption value; a second evaluation result is obtained according to the second prediction result and a second actual oil consumption value; a target evaluation result is obtained according to the first evaluation result and the second evaluation result, and is output.

[0203] Optionally, the fusing and processing of the multi-source heterogeneous data to obtain the target data comprises:

[0204] Data cleaning is performed on the multi-source heterogeneous data to obtain first intermediate data;

[0205] Format conversion is performed on the first intermediate data to obtain second intermediate data;

[0206] Data fusion processing is performed on the second intermediate data to obtain target data.

[0207] Optionally, the training process of the first fuel consumption prediction model comprises:

[0208] A first data set is obtained, and the first data set comprises a first training set and a first verification set;

[0209] Data in the first training set is input into an input layer of a preset machine learning model for processing to obtain a first output;

[0210] The first output is input into a loop processing layer of the preset machine learning model for processing to obtain a second output;

[0211] The second output is input into an output layer of the preset machine learning model for processing to obtain a first prediction result;

[0212] The preset machine learning model is tuned according to the first prediction result, the first verification set and a mean absolute percentage error to obtain the first fuel consumption prediction model.

[0213] Optionally, the training process of the second fuel consumption prediction model comprises:

[0214] A second data set is obtained, and the second data set comprises a second training set and a second verification set;

[0215] Data in the second training set is input into an input layer of a preset machine learning model for processing to obtain a third output;

[0216] The third output is input into a loop processing layer of the preset machine learning model for processing to obtain a fourth output;

[0217] The fourth output is input into an output layer of the preset machine learning model for processing to obtain a second prediction result;

[0218] The preset machine learning model is tuned according to the second prediction result, the second verification set and a mean absolute percentage error to obtain the second fuel consumption prediction model.

[0219] Optionally, the first data is input into the second fuel consumption prediction model for fuel consumption prediction to obtain the first prediction result, which comprises:

[0220] The first data is input into an input layer of the second fuel consumption prediction model for processing to obtain a third output;

[0221] inputting the third output into a loop processing layer of the preset machine learning model to obtain a fourth output;

[0222] inputting the fourth output into an output layer of the preset machine learning model to obtain a first prediction result.

[0223] Optionally, the second data is input into a first fuel consumption prediction model to obtain a second prediction result, including:

[0224] inputting the second data into an input layer of the first fuel consumption prediction model to obtain a first output;

[0225] inputting the first output into a loop processing layer of the preset machine learning model to obtain a second output;

[0226] inputting the second output into an output layer of the preset machine learning model to obtain a second prediction result.

[0227] Optionally, a first evaluation result is obtained according to the first prediction result and a first actual fuel consumption value, including:

[0228] a first evaluation result is obtained according to the first predicted fuel consumption, the first actual fuel consumption and a first mean absolute percentage error;

[0229] a second evaluation result is obtained according to the second prediction result and a second fuel consumption value, including:

[0230] a second evaluation result is obtained according to the second predicted fuel consumption, the second actual fuel consumption and a second mean absolute percentage error.

[0231] Optionally, a target evaluation result is obtained according to the first evaluation result and the second evaluation result, including:

[0232] an intermediate evaluation result is obtained according to the first evaluation result and the second evaluation result;

[0233] a target evaluation result is obtained according to the intermediate evaluation result and an index weight.

[0234] It should be noted that all implementation manners in the method embodiments are applicable to the device embodiments, and can achieve the same technical effects.

[0235] The embodiments of the present application also provide a computing device readable storage medium, the computing device readable storage medium stores a program, and the program is executed by a processor to implement the ship energy efficiency measure application effect evaluation method. All implementation manners in the method embodiments are applicable to the computing device readable storage medium embodiments, and can achieve the same technical effects.

[0236] Those skilled in the art can clearly understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0237] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working processes of the above-described system, device and unit can refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

[0238] In the embodiments provided by the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. For example, the apparatus embodiments described above are merely schematic, for example, the division of the units is merely a logical function division, and another division manner can be used in actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be omitted or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or in other forms.

[0239] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0240] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically independently, or two or more units can be integrated into one unit.

[0241] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the parts that contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a ROM, a RAM, a magnetic disk or an optical disk, and various program code storage media.

[0242] In addition, it should be noted that in the device and method of the present application, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombination should be considered as equivalent solutions of the present application. Moreover, the steps of performing the above series of processes can naturally be executed in time sequence according to the order of description, but do not necessarily have to be executed in time sequence, and some steps can be executed in parallel or independently of each other. It can be understood by those skilled in the art that all or any steps or components of the method and device of the present application can be implemented in hardware, firmware, software or a combination thereof in any computing device (including processors, storage media, etc.) or network of computing devices, which can be implemented by those skilled in the art using their basic programming skills after reading the description of the present application.

[0243] Therefore, the object of the present application can also be achieved by running a program or a set of programs on any computing device. The computing device can be a commonly known general-purpose device. Therefore, the object of the present application can also be achieved by providing a program product containing program code for implementing the method or device. That is, such a program product also constitutes the present application, and a storage medium storing such a program product also constitutes the present application. Obviously, the storage medium can be any commonly known storage medium or any storage medium developed in the future. It should be noted that in the device and method of the present application, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombination should be considered as equivalent solutions of the present application. Moreover, the steps of performing the above series of processes can naturally be executed in time sequence according to the order of description, but do not necessarily have to be executed in time sequence. Some steps can be executed in parallel or independently of each other.

[0244] The above is the preferred embodiment of the present application. It should be noted that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which should also be considered within the scope of protection of the present application.

Claims

1. A method for evaluating the application effect of ship energy efficiency measures, characterized in that, The method comprises the following steps: acquiring multi-source heterogeneous data of a ship; performing fusion processing on the multi-source heterogeneous data to obtain target data; dividing the target data into first data before application of energy efficiency measures and second data after application of energy efficiency measures according to time of application of energy efficiency measures; inputting the first data into a second oil consumption prediction model to perform oil consumption prediction and obtain a first prediction result; the second oil consumption prediction model is obtained by training a preset machine learning model according to a second data set; inputting the second data into a first oil consumption prediction model to perform oil consumption prediction and obtain a second prediction result; the first oil consumption prediction model is obtained by training a preset machine learning model according to a first data set; obtaining a first evaluation result according to the first prediction result and a first actual oil consumption value; obtaining a second evaluation result according to the second prediction result and a second actual oil consumption value; obtaining a target evaluation result according to the first evaluation result and the second evaluation result, and outputting the target evaluation result; wherein the first oil consumption prediction model is an oil consumption prediction model trained by data before application of energy efficiency measures, and a training process of the first oil consumption prediction model comprises: acquiring a first data set, the first data set being objective data accumulated historically before application of energy efficiency measures, and the first data set comprising a first training set and a first verification set; inputting data in the first training set into an input layer of a preset machine learning model to perform processing and obtain a first output; inputting the first output into a loop processing layer of the preset machine learning model to perform processing and obtain a second output; inputting the second output into an output layer of the preset machine learning model to perform processing and obtain a first prediction result; optimizing the preset machine learning model according to the first prediction result, the first verification set and a mean absolute percentage error to obtain the first oil consumption prediction model; the second oil consumption prediction model is an oil consumption prediction model trained by data after application of energy efficiency measures, and a training process of the second oil consumption prediction model comprises: acquiring a second data set, the second data set being objective data accumulated historically after application of energy efficiency measures, and the second data set comprising a second training set and a second verification set; inputting data in the second training set into an input layer of a preset machine learning model to perform processing and obtain a third output; inputting the third output into a loop processing layer of the preset machine learning model to perform processing and obtain a fourth output; inputting the fourth output into an output layer of the preset machine learning model to perform processing and obtain a second prediction result; optimizing the preset machine learning model according to the second prediction result, the second verification set and a mean absolute percentage error to obtain the second oil consumption prediction model.

2. The method of evaluating the application effect of ship energy efficiency measures according to claim 1, characterized in that, The fusion processing on the multi-source heterogeneous data to obtain target data comprises: performing data cleaning on the multi-source heterogeneous data to obtain first intermediate data; performing format conversion on the first intermediate data to obtain second intermediate data; performing data fusion processing on the second intermediate data to obtain target data.

3. The method of evaluating the application effect of ship energy efficiency measures according to claim 1, characterized in that, The oil consumption prediction by inputting the first data into a second oil consumption prediction model to obtain a first prediction result comprises: The first data is input into an input layer of the second fuel consumption prediction model for processing to obtain a third output; The third output is input into a loop processing layer of the preset machine learning model for processing to obtain a fourth output; The fourth output is input into an output layer of the preset machine learning model for processing to obtain a first prediction result.

4. The method of evaluating the application effect of ship energy efficiency measures according to claim 1, characterized in that, The second data is input into the first fuel consumption prediction model for fuel consumption prediction to obtain a second prediction result, including: The second data is input into an input layer of the first fuel consumption prediction model for processing to obtain a first output; The first output is input into a loop processing layer of the preset machine learning model for processing to obtain a second output; The second output is input into an output layer of the preset machine learning model for processing to obtain a second prediction result.

5. The method of evaluating the application effect of ship energy efficiency measures according to claim 1, characterized in that, According to the first prediction result and a first actual fuel consumption value, a first evaluation result is obtained, including: According to the first predicted fuel consumption, the first actual fuel consumption, and a first mean absolute percentage error, the first evaluation result is obtained; According to the second prediction result and a second fuel consumption value, a second evaluation result is obtained, including: According to the second predicted fuel consumption, the second actual fuel consumption, and a second mean absolute percentage error, the second evaluation result is obtained.

6. The method of evaluating the application effect of ship energy efficiency measures according to claim 1, characterized in that, According to the first evaluation result and the second evaluation result, a target evaluation result is obtained, including: According to the first evaluation result and the second evaluation result, an intermediate evaluation result is obtained; According to the intermediate evaluation result and an index weight, the target evaluation result is obtained.

7. An apparatus for evaluating the application effect of a ship energy efficiency measure, characterized by, including: An acquisition module is configured to acquire ship multi-source heterogeneous data; A processing module is configured to perform fusion processing on the multi-source heterogeneous data to obtain target data; The target data is divided into first data before application of energy efficiency measures and second data after application of energy efficiency measures according to time of application of energy efficiency measures; the first data is input into a second fuel consumption prediction model for fuel consumption prediction to obtain a first prediction result; the second fuel consumption prediction model is obtained by training a preset machine learning model according to a second data set; The second data is input into a first fuel consumption prediction model for fuel consumption prediction to obtain a second prediction result; the first fuel consumption prediction model is obtained by training a preset machine learning model according to a first data set; According to the first prediction result and a first actual fuel consumption value, a first evaluation result is obtained; According to the second prediction result and a second actual fuel consumption value, a second evaluation result is obtained; According to the first evaluation result and the second evaluation result, a target evaluation result is obtained and output; The first fuel consumption prediction model is a fuel consumption prediction model trained by data before application of energy efficiency measures, and a training process of the first fuel consumption prediction model includes: A first data set is acquired, the first data set is objective data before application of energy efficiency measures accumulated historically, and the first data set includes a first training set and a first verification set; Data in the first training set is input into an input layer of a preset machine learning model for processing to obtain a first output; The first output is input into a loop processing layer of the preset machine learning model for processing to obtain a second output; The second output is input into an output layer of the preset machine learning model for processing to obtain a first prediction result; According to the first prediction result, the first verification set and the mean absolute percentage error, a preset machine learning model is optimized to obtain a first fuel consumption prediction model; The second fuel consumption prediction model is a fuel consumption prediction model trained by data after application of energy efficiency measures, and a training process of the second fuel consumption prediction model comprises: A second data set is obtained, the second data set is objective data accumulated historically after application of energy efficiency measures, and the second data set comprises a second training set and a second verification set; Data in the second training set is input into an input layer of a preset machine learning model for processing to obtain a third output; The third output is input into a recurrent processing layer of the preset machine learning model for processing to obtain a fourth output; The fourth output is input into an output layer of the preset machine learning model for processing to obtain a second prediction result; According to the second prediction result, the second verification set and the mean absolute percentage error, the preset machine learning model is optimized to obtain the second fuel consumption prediction model.

8. A computing device readable storage medium characterized by, The computing device readable storage medium stores a program which, when executed by a processor, implements the method of any one of claims 1 to 6.

Images