An intelligent control system for connecting power supply and preventing reverse current

Abstract

The present invention belongs to the technical field of electric power control, and discloses an intelligent control system for connecting power supply and preventing reverse current; the system comprises: a data acquisition module for acquiring historical electric power data; a data prediction module for analyzing historical electric power data and predicting future electric power data; a reverse current detection module for analyzing future electric power data and judging whether to generate a reverse current instruction; a reverse current analysis module for analyzing the reverse current type based on future electric power data if a reverse current instruction is generated; a dynamic adjustment module for integrating the reverse current type and future electric power data and formulating a dynamic adjustment strategy; a reverse current early warning module for executing the dynamic adjustment strategy and judging whether to generate an early warning instruction; the present invention can effectively monitor and prevent reverse current phenomenon and improve the safety and stability of the shore power system; it not only optimizes the power supply management and reduces the risk of equipment damage, but also improves the overall operation efficiency and safety and reliability of the power system.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power control, and more specifically to an intelligent control system for connecting power supply and preventing reverse electricity. Background Art

[0002] With the increasing global attention paid to environmental protection and sustainable development, the application of shore power equipment in transportation hubs such as ports has gradually become popular. Shore power technology provides power supply to docked ships, replacing traditional fuel power generation, significantly reducing emissions and noise, and improving the surrounding environment. However, the shore power system also faces challenges in the stability and safety of power supply in actual operation. One of the most prominent problems is the reverse current phenomenon. In the power supply system connected to the shore power equipment, reverse current usually occurs when the docked equipment fails to effectively manage the flow of current, especially when multiple ships are connected to the shore power at the same time. The direction of current flow may be confused, causing power to flow back to the power grid. It will not only damage the power equipment, but also affect the stability of the power grid. Therefore, it is urgent to develop an intelligent control system to prevent the occurrence of reverse current.

[0003] The patent with announcement number CN103475028B discloses a reverse current prevention control system and control method for a power supply system using a braking feedback device; the invention comprises: a reverse current prevention monitoring device for detecting the direction of current, which is arranged at the connection between the distribution transformer and the medium voltage ring network of the power supply system and is located at the outlet of the distribution transformer; the reverse current prevention monitoring device is communicatively connected to the SCADA monitoring system; the SCADA monitoring system is communicatively connected to each braking feedback converter; the invention utilizes the existing system structure, and sets a reverse current prevention monitoring device at each reverse current prevention monitoring point to detect the direction of current on the distribution transformer side at the monitoring point in real time; when reverse current is detected, the reverse current signal is sent to each subway braking feedback device through the SCADA monitoring system to control it to reduce the feedback output power to the medium voltage ring network until the reverse current signal disappears; the reverse current of the power supply system can be prevented from occurring, and the safe and reliable operation of the high voltage power grid can be avoided from being affected.

[0004] However, the above technology mainly relies on the anti-backflow monitoring device installed between the distribution transformer and the medium-voltage ring network for anti-backflow monitoring. The monitoring range is narrow and cannot fully capture the reverse current situation under multi-point access. In addition, the above technology only eliminates the reverse flow phenomenon by reducing the output power, lacks flexibility, and cannot eliminate the reverse flow phenomenon between multiple points, which leads to unstable power supply and affects the normal operation of equipment.

[0005] In view of this, the present invention proposes a connection power supply reverse power prevention intelligent control system to solve the above problem. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: an intelligent control system for connecting power supply and preventing reverse power, comprising:

[0007] Data acquisition module, used to collect historical power data;

[0008] Data prediction module, used to analyze historical power data and predict future power data;

[0009] The reverse power detection module is used to analyze future power data and determine whether to generate a reverse power instruction;

[0010] A reverse power analysis module is used to analyze the reverse power type based on future power data if a reverse power instruction is generated;

[0011] Dynamic adjustment module, used to integrate reverse power type and future power data to formulate dynamic adjustment strategies;

[0012] The reverse current warning module is used to execute dynamic adjustment strategies and determine whether to generate a warning instruction.

[0013] Further, the historical power data is power data at historical moments, the historical power data includes n groups of power data, n is an integer greater than 1, and one group of power data corresponds to one moment; the power data includes shore power data and ship data; the shore power data includes shore power voltage and shore power current; the ship data includes ship voltage, ship current and ship load, and the ship data corresponds one-to-one to the ship connected to the shore power equipment;

[0014] The method for predicting future power data comprises:

[0015] Collect real-time environmental data and future environmental data; real-time environmental data refers to the environmental data at the current moment, and future environmental data refers to the future Environmental data at moments, m is an integer greater than 2, and the environmental data includes environmental temperature, environmental humidity and extreme weather data; different digital labels are set for different extreme weather data, and marked as weather labels; the extreme weather data in the environmental data is replaced with the corresponding weather labels, the historical power data and the real-time environmental data are used as analysis data, the analysis data is input into the trained power forecasting model, the power data at the next moment is predicted, and marked as prediction data; the earliest power data at the corresponding moment in the historical power data is replaced with the prediction data, and the historical power data is marked as replacement power data; the historical power data in the analysis data is replaced with replacement power data, and the real-time environmental data is replaced with the environmental data in the future environmental data that is the same as the corresponding moment of the prediction data predicted last time, and the real-time environmental data is input into the trained power forecasting model again, and the power data at the next moment of the corresponding moment of the prediction data predicted last time is predicted is predicted; in this cycle, a total of m groups of prediction data are predicted, and the m groups of prediction data correspond to m moments in the future, and the m groups of prediction data are used as future power data.

[0016] Furthermore, the power prediction model includes a shore power voltage prediction model, a shore power current prediction model, a ship voltage prediction model, a ship current prediction model and a ship load prediction model; the power prediction model corresponds to the power data one by one, and the training process of each model in the power prediction model is consistent, and all are deep neural network models;

[0017] The training process of the shore power voltage prediction model includes:

[0018] P groups of analysis data are collected in advance, corresponding shore power voltages are set for the p groups of analysis data, where p is an integer greater than 1, and the analysis data and the corresponding shore power voltage are converted into a corresponding set of feature vectors; each set of feature vectors is used as the input of a shore power voltage prediction model, and the shore power voltage prediction model uses a set of predicted shore power voltages corresponding to each set of analysis data as output, and uses the actual shore power voltage corresponding to each set of analysis data as a prediction target, where the actual shore power voltage is the pre-set shore power voltage corresponding to the analysis data; minimizing the sum of prediction errors of all analysis data is used as a training target; and the shore power voltage prediction model is trained until the sum of prediction errors converges and the training is stopped.

[0019] Furthermore, the method for determining whether to generate a reverse power instruction includes:

[0020] According to the future power data, calculate the current change and load change at the next m moments; use the current change and load change at the next m moments and the shore power current in the future power data as test data; build a detection forest according to the test data; calculate the actual distance and expected distance of each data in the future data according to the detection forest; calculate the abnormal value of each data in the future data according to the actual distance and expected distance of each data in the future data; preset an abnormal threshold; if the abnormal value of the data in the future data is greater than or equal to the abnormal threshold, generate a reverse power instruction; if the abnormal value of each data in the future data is less than the abnormal threshold, do not generate a reverse power instruction;

[0021] The method for calculating the current change amount and the load change amount at the next m moments includes:

[0022] The latest power data at the corresponding moment in the historical power data is obtained and marked as real-time power data; the ship current in the real-time power data and the ship current in the future power data are sorted from late to early according to the corresponding moment; each ship current is subtracted from the ship current ranked next in positive order to obtain the current change at the next m moments; the ship load in the real-time power data and the ship load in the future power data are sorted from late to early according to the corresponding moment; each ship load is subtracted from the ship load ranked next in positive order to obtain the load change at the next m moments.

[0023] Furthermore, the step of constructing the detection forest includes:

[0024] Step A1: according to the test data, obtaining a data range value, the data range value includes a current variation range value, a load variation range value and a shore power current range value;

[0025] Step A2: Randomly select a type of data from the test data and mark it as selected data;

[0026] Step A3: randomly select a value in the data range corresponding to the selected data as a split point, and split the selected data according to the split point to obtain a split sub-set;

[0027] Step A4: Obtain a numerical range for each slicing sub-set, wherein the maximum value of the numerical range is the maximum value of the numerical range in the slicing sub-set, and the minimum value of the numerical range is the minimum value of the numerical range in the slicing sub-set; randomly select a slicing point in the numerical range corresponding to each slicing sub-set, and obtain a total of q slicing points, where q is the number of slicing sub-sets; and slicing each slicing sub-set according to the q slicing points;

[0028] Step A5: loop step A4 until there is only one value in each segmentation sub-set, the loop ends, obtains the detection tree corresponding to the selected data, and proceeds to step A6;

[0029] Step A6: looping steps A2 to A5 until each type of data in the test data is marked as selected data, the loop ends, and the detection trees corresponding to each selected data are combined into a detection forest;

[0030] Methods for calculating actual distance and expected distance include:

[0031] The actual distance is calculated as follows: the number of times the data is split in each corresponding detection tree is counted and marked as the number of splits. Each number of splits is added up in sequence and then divided by the number of detection trees corresponding to the data to obtain the actual distance corresponding to the data.

[0032] The expected distance is calculated as: ; In the formula, is the expected distance;

[0033] Methods for calculating outliers include: ; In the formula, is an outlier, is the actual distance.

[0034] Further, the reverse power types include inter-ship reverse power and shore power reverse power;

[0035] The method for analyzing the reverse electricity type includes:

[0036] Obtain data in the future data corresponding to the reverse power instruction and mark it as abnormal data; obtain the predicted data corresponding to the moment corresponding to the abnormal data and mark it as abnormal predicted data; obtain the shore power voltage and ship voltage in the abnormal predicted data; compare the shore power voltage with each ship voltage respectively; if the shore power voltage is greater than or equal to each ship voltage, analyze the reverse power type as inter-ship reverse power; if there is a ship voltage greater than the shore power voltage, analyze the reverse power type as shore power reverse power, and mark the ship corresponding to the ship voltage as an abnormal ship.

[0037] Furthermore, the step of formulating a dynamic adjustment strategy includes:

[0038] Step B1: If the reverse power type is inter-ship reverse power, proceed to step B2; if the reverse power type is shore power reverse power, proceed to step B9;

[0039] Step B2: Preset initialization temperature , minimum temperature And the maximum number of iterations , and let the current temperature ;

[0040] Step B3: Construct M groups of adjustment parameters, where M is an integer greater than 1; randomly set a feasible solution , a feasible solution To adjust the parameters, the feasible solution The range is M group parameters;

[0041] Step B4: Determine the optimization function and cooling coefficient ;

[0042] Step B5: Calculate feasible solutions The corresponding optimization degree ; with feasible solutions For the current point, perform random perturbations in the neighborhood of the current point to obtain a new feasible solution , and calculate the new feasible solution The corresponding optimization degree ;

[0043] Step B6: Calculate the optimization difference , optimization difference The expression is ; If the optimization difference , then let ; If the optimization difference , then calculate the probability , according to the probability make ; Probability The expression is: ; In the formula, e is a natural constant;

[0044] Step B7: Repeat steps B5 to B6 until the number of loops reaches the maximum number of iterations. When the cycle ends, let the current temperature ; Set the maximum number of iterations ;

[0045] Step B8: Repeat steps B5 to B7 until the current temperature reaches When , the loop ends and a feasible solution is obtained The corresponding adjustment parameters;

[0046] Step B9: Subtract the shore power voltage from the abnormal ship's ship voltage and multiply by k to obtain the voltage adjustment value. .

[0047] Furthermore, in step B3, the method for constructing M groups of adjustment parameters is: obtaining the shore power current in the abnormal prediction data; dividing the shore power current into N values, the sum of the N values ​​is the shore power current, and N is the number of ships connected to the shore power equipment; taking the N values ​​as a group of adjustment parameters, constructing a total of M groups of adjustment parameters, the M groups of adjustment parameters are inconsistent, and the values ​​in the adjustment parameters correspond to the ships one by one;

[0048] In step B4, the expression of the optimization function is: ;

[0049] In the formula, To optimize, is the reverse electrical probability;

[0050] The method to obtain the reverse electrical probability is: The corresponding adjustment parameters and abnormal prediction data are used as research data. The research data are input into the trained probability prediction model to predict the corresponding reverse power probability. The training process of the probability prediction model is consistent with the training process of the shore power voltage prediction model, and both are deep neural network models.

[0051] Furthermore, the temperature reduction coefficient is determined The steps include:

[0052] Step C1: Preset initial coefficient search interval ; Set the split ratio ;

[0053] Step C2: According to the segmentation ratio , the initial coefficient search interval Divide two new points and , , ; Construct an optimization set, which includes different cooling coefficients And the corresponding optimization degree;

[0054] Step C3: Obtain according to the optimization degree set and The corresponding optimization degree will be The corresponding optimization degree is marked as the first optimization degree. The corresponding optimization degree is marked as the second optimization degree;

[0055] Step C4: if the first optimization degree is greater than or equal to the second optimization degree, the minimum value of the updated coefficient search interval is the point corresponding to the first optimization degree, and the maximum value is the maximum value of the coefficient search interval before updating; if the first optimization degree is less than the second optimization degree, the minimum value of the updated coefficient search interval is the minimum value of the coefficient search interval before updating, and the maximum value is the point corresponding to the second optimization degree;

[0056] Step C5: preset a width threshold, subtract the minimum value from the maximum value in the updated coefficient search interval to obtain the interval width; compare the interval width with the width threshold; if the interval width is greater than or equal to the width threshold, the updated coefficient search area is divided according to the segmentation ratio , re-dividing two new points and , and return to step C3; if the interval width is less than the width threshold, the average of the maximum and minimum values ​​in the updated coefficient search area is used as the cooling coefficient .

[0057] Further, when the time corresponding to the abnormal prediction data is reached, a dynamic adjustment strategy is executed;

[0058] The method for determining whether to generate an early warning instruction includes:

[0059] If the reverse power type is reverse flow between ships, the current flow direction of the abnormal ship is collected and marked as the ship flow direction value; if the ship flow direction value is a positive value, no warning instruction is generated; if the ship flow direction value is a negative value, a warning instruction is generated; if the reverse power type is shore power reverse flow, the current flow direction of the shore power equipment is collected and marked as the shore current direction value; if the shore current direction value is a positive value, no warning instruction is generated; if the shore current direction value is a negative value, a warning instruction is generated.

[0060] The technical effects and advantages of the intelligent control system for connecting power supply and preventing reverse electricity of the present invention are as follows:

[0061] By collecting historical power data and using deep learning technology to predict future power data, we can accurately detect reverse current phenomena and identify reverse current types, so as to formulate targeted dynamic adjustment strategies; and after the dynamic adjustment strategy is executed, the execution of the adjustment strategy is judged in real time, and reverse current warnings are issued in time; through the collaborative work of modules such as data collection, predictive analysis, reverse current detection and dynamic adjustment, it can effectively monitor and prevent current reverse flow phenomena and improve the safety and stability of the shore power system; it not only optimizes power supply management and reduces the risk of equipment damage, but also improves the overall operating efficiency and safety and reliability of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a schematic diagram of an intelligent control system for connecting power supply and preventing reverse current according to Embodiment 1 of the present invention;

[0063] Figure 2 This is a flow chart of the detection forest construction method of Example 1 of the present invention. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] Example 1

[0066] See also Figure 1As shown, the embodiment of the invention provides a connected power supply reverse current prevention intelligent control system, including a data acquisition module, a data prediction module, a reverse current detection module, a reverse current analysis module, a dynamic adjustment module and a reverse current warning module; each module is connected by wire and / or wireless means to achieve data transmission between modules;

[0067] Data acquisition module, used to collect historical power data.

[0068] The historical power data is the power data at historical moments, and the historical power data includes n groups of power data, where n is an integer greater than 1, and one group of power data corresponds to one moment; the power data includes shore power data and ship data; the shore power data includes shore power voltage and shore power current; the ship data includes ship voltage, ship current and ship load, and the ship data corresponds one-to-one to the ship connected to the shore power equipment; the shore power voltage is obtained by the voltage transformer installed at the output end of the shore power equipment, the shore power current is obtained by the current transformer installed at the output end of the shore power equipment, the ship voltage is obtained by the voltage transformer installed at the main power input end of the ship's power system, the ship current is obtained by the current transformer installed at the main power input end of the ship's power system, and the ship load is obtained by the power sensor installed on the main load distribution circuit of the ship.

[0069] The shore power voltage reflects the voltage level supplied by the shore power equipment to the ship, and the ship voltage reflects the voltage level of the ship's power system. Under normal circumstances, the shore power voltage should be higher than the ship voltage, so that the current flows from the shore power equipment to the ship to ensure power supply. When the ship voltage is higher than the shore power voltage, the ship's current will flow back to the shore power equipment, resulting in reverse current flow. The shore power current indicates the current provided by the shore power equipment to the ship. When the shore power current is negative, that is, the current flows in the opposite direction, indicating that reverse current flow has occurred. The ship current reflects the current consumed by the ship, and the ship load indicates the power load of the ship's power system. When the ship current suddenly drops, it means that the ship load has dropped and the ship's power demand has dropped, which will cause the ship voltage to rise, thereby causing reverse current flow.

[0070] The data prediction module is used to analyze historical power data and predict future power data.

[0071] Methods for predicting future electricity data include:

[0072] Collect real-time environmental data and future environmental data; real-time environmental data refers to the environmental data at the current moment, and future environmental data refers to the future The environmental data at each moment is obtained from the website of the National Meteorological Administration. Different digital labels are set for different extreme weather data and marked as weather labels. For example, the digital label for normal weather is 0, the digital label for storm is 1, and the digital label for typhoon is 2. The extreme weather data in the environmental data is replaced with the corresponding weather labels. The historical power data and the real-time environmental data are used as analysis data. The analysis data is input into the trained power forecasting model to predict the power at the next moment. Data is obtained and marked as predicted data; the earliest power data at the corresponding moment in the historical power data is replaced with the predicted data, and the historical power data is marked as the replacement power data; the historical power data in the analysis data is replaced with the replacement power data, and the real-time environmental data is replaced with the environmental data in the future environmental data that is the same as the corresponding moment of the predicted data predicted last time, and is input into the trained power prediction model again to predict the power data at the next moment corresponding to the predicted data predicted last time; in this cycle, a total of m groups of predicted data are predicted, and the m groups of predicted data correspond to m moments in the future, and the m groups of predicted data are used as future power data.

[0073] The power prediction model includes shore power voltage prediction model, shore power current prediction model, ship voltage prediction model, ship current prediction model and ship load prediction model; the power prediction model corresponds to the power data one by one, and the training process of each model in the power prediction model is consistent, all of which are deep neural network models; the training process of the shore power voltage prediction model includes:

[0074] Collect p groups of analysis data in advance, set corresponding shore power voltages for the p groups of analysis data, where p is an integer greater than 1, and convert the analysis data and the corresponding shore power voltages into a corresponding set of feature vectors; the shore power voltages corresponding to the analysis data are collected by technicians skilled in the art during the historical power data prediction process, and the shore power voltages at the next moment for each group of analysis data are collected, and the corresponding shore power voltages are set for the p groups of analysis data in sequence;

[0075] Each set of feature vectors is used as the input of the shore power voltage prediction model. The shore power voltage prediction model uses a set of predicted shore power voltages corresponding to each set of analysis data as output, and uses the actual shore power voltage corresponding to each set of analysis data as the prediction target. The actual shore power voltage is the pre-set shore power voltage corresponding to the analysis data. Minimizing the sum of the prediction errors of all analysis data is used as the training target. The calculation formula of the prediction error is: ,in is the prediction error, w is the group number of the eigenvector corresponding to the analysis data, is the predicted shore power voltage corresponding to the wth group of analysis data, is the actual shore power voltage corresponding to the wth group of analysis data; the shore power voltage prediction model is trained until the sum of the prediction errors reaches convergence and the training is stopped.

[0076] The above-mentioned shore power voltage prediction model is specifically a deep neural network model; it includes an input layer, a hidden layer and an output layer; each hidden layer includes multiple neurons, each neuron is connected to the neurons in the next layer, and the connection contains weights, which determine the importance and influence of data transmission in the neural network; an activation function is applied to each neuron between the hidden layer and the output layer, and the activation function introduces nonlinearity, allowing the network to learn more complex patterns and features.

[0077] It should be understood that environmental data will indirectly or directly affect power data by affecting the performance of power equipment, the transmission efficiency of current and voltage, load demand and system stability; therefore, when predicting future power data, environmental factors need to be taken into consideration to ensure accurate prediction of power data at future times.

[0078] The reverse power detection module is used to analyze future power data and determine whether to generate a reverse power instruction.

[0079] The method for determining whether to generate a reverse power instruction includes:

[0080] According to the future power data, the current change and load change at the next m moments are calculated; the current change and load change at the next m moments and the shore power current in the future power data are used as test data; according to the test data, a detection forest is constructed; according to the detection forest, the actual distance and expected distance of each data in the future data are calculated; according to the actual distance and expected distance of each data in the future data, the abnormal value of each data in the future data is calculated; an abnormal threshold is preset, and the abnormal threshold is preset by technical personnel in this field according to actual conditions; if there is an abnormal value of data in the future data that is greater than or equal to the abnormal threshold, a reverse power instruction is generated; if the abnormal value of each data in the future data is less than the abnormal threshold, no reverse power instruction is generated.

[0081] The method for calculating the current change and load change at the next m moments includes:

[0082] The latest power data at the corresponding moment in the historical power data is obtained and marked as real-time power data; the ship current in the real-time power data and the ship current in the future power data are sorted from late to early according to the corresponding moment; each ship current is subtracted from the ship current ranked next in positive order to obtain the current change at the next m moments; the ship load in the real-time power data and the ship load in the future power data are sorted from late to early according to the corresponding moment; each ship load is subtracted from the ship load ranked next in positive order to obtain the load change at the next m moments.

[0083] like Figure 2 As shown, the steps to build a detection forest include:

[0084] Step A1: According to the test data, a data range value is obtained, the data range value including a current variation range value, a load variation range value and a shore power current range value; the maximum value of the current variation range value is the maximum value of the current variation in the test data, and the minimum value of the current variation range value is the minimum value of the current variation in the test data; the maximum value of the load variation range value is the maximum value of the load variation in the test data, and the minimum value of the load variation range value is the minimum value of the load variation in the test data; the maximum value of the shore power current range value is the maximum value of the shore power current in the test data, and the minimum value of the shore power current range value is the minimum value of the shore power current in the test data;

[0085] Step A2: Randomly select a type of data from the test data and mark it as selected data;

[0086] Step A3: randomly select a value in the data range corresponding to the selected data as a split point, and split the selected data according to the split point to obtain a split sub-set;

[0087] Step A4: Obtain a numerical range for each slicing sub-set, wherein the maximum value of the numerical range is the maximum value of the numerical range in the slicing sub-set, and the minimum value of the numerical range is the minimum value of the numerical range in the slicing sub-set; randomly select a slicing point in the numerical range corresponding to each slicing sub-set, and obtain a total of q slicing points, where q is the number of slicing sub-sets; and slicing each slicing sub-set according to the q slicing points;

[0088] Step A5: loop step A4 until there is only one value in each segmentation sub-set, the loop ends, obtains the detection tree corresponding to the selected data, and proceeds to step A6;

[0089] Step A6: looping steps A2 to A5 until each type of data in the test data is marked as selected data, the loop ends, and the detection trees corresponding to each selected data are combined into a detection forest.

[0090] Methods for calculating actual distance and expected distance include:

[0091] The calculation method of the actual distance is: statistical data is divided into each corresponding detection tree times, and marked as the number of divisions, each division number is added in sequence, and then divided by the number of detection trees corresponding to the data, as the actual distance corresponding to the data; illustratively, the data includes 2, 4, 6, and 9, and the division point is selected as 3. At this time, one division sub-set is 2, and the other division sub-set is 4, 6, and 9; the division point is selected as 7, at this time, one division sub-set is 4 and 6, and the other division sub-set is 9; the division point is selected as 5, at this time, one division sub-set is 4, and the other division sub-set is 6; therefore, the number of divisions corresponding to 2 is 1, the number of divisions corresponding to 4 and 6 is 3, and the number of divisions corresponding to 9 is 2.

[0092] The expected distance is calculated as: ; In the formula, is the expected distance.

[0093] Methods for calculating outliers include: ; In the formula, is an outlier, is the actual distance.

[0094] The reverse power analysis module is used to analyze the reverse power type based on future power data if a reverse power instruction is generated.

[0095] The types of reverse power flow include inter-ship reverse power flow and shore power reverse power flow. Inter-ship reverse power flow refers to the situation where the current flows from one ship to another when multiple ships are connected to shore power equipment. Shore power reverse power flow refers to the phenomenon where the current flows from the ship to the shore power equipment when the ship is connected to the shore power equipment.

[0096] Methods for analyzing reverse electrical types include:

[0097] Obtain data in the future data corresponding to the reverse power instruction and mark it as abnormal data; obtain the predicted data corresponding to the moment corresponding to the abnormal data and mark it as abnormal predicted data; obtain the shore power voltage and ship voltage in the abnormal predicted data; compare the shore power voltage with each ship voltage respectively; if the shore power voltage is greater than or equal to each ship voltage, analyze the reverse power type as inter-ship reverse power; if there is a ship voltage greater than the shore power voltage, analyze the reverse power type as shore power reverse power, and mark the ship corresponding to the ship voltage as an abnormal ship.

[0098] The dynamic adjustment module is used to integrate reverse power type and future power data to formulate dynamic adjustment strategies.

[0099] The steps to developing a dynamic adjustment strategy include:

[0100] Step B1: If the reverse power type is inter-ship reverse power, proceed to step B2; if the reverse power type is shore power reverse power, proceed to step B9;

[0101] Step B2: Preset initialization temperature , minimum temperature And the maximum number of iterations , and let the current temperature ;

[0102] Step B3: Construct M groups of adjustment parameters, where M is an integer greater than 1; randomly set a feasible solution , a feasible solution To adjust the parameters, the feasible solution The range is M group parameters;

[0103] Step B4: Determine the optimization function and cooling coefficient ;

[0104] Step B5: Calculate feasible solutions The corresponding optimization degree ; with feasible solutions For the current point, perform random perturbations in the neighborhood of the current point to obtain a new feasible solution , and calculate the new feasible solution The corresponding optimization degree ;

[0105] Step B6: Calculate the optimization difference , optimization difference The expression is ; If the optimization difference , then let , the new feasible solution The value of is given to the feasible solution ; If the optimization difference , then calculate the probability , according to the probability make ; Probability The expression is: ; In the formula, e is a natural constant;

[0106] Step B7: Repeat steps B5 to B6 until the number of loops reaches the maximum number of iterations. When the cycle ends, let the current temperature , that is, cool down the current temperature in step B2, and assign the value after cooling to the current temperature; let the maximum number of iterations , that is, assign the value of the reduced maximum number of iterations to the maximum number of iterations; if the reduced maximum number of iterations is not an integer, the reduced maximum number of iterations is rounded up to make it an integer;

[0107] Step B8: Repeat steps B5 to B7 until the current temperature reaches When , the loop ends and a feasible solution is obtained The corresponding adjustment parameters;

[0108] Step B9: Subtract the shore power voltage from the abnormal ship's ship voltage and multiply by k to obtain the voltage adjustment value. .

[0109] In the above step B2, the initialization temperature , minimum temperature , Temperature reduction coefficient And the maximum number of iterations As preset parameters, the preset parameters are collected by a technician in this field during the formulation of the historical dynamic adjustment strategy. v groups of different abnormal prediction data are collected, and multiple groups of different preset parameters are pre-set for one group of abnormal prediction data in turn. The adjustment parameters are obtained according to steps B2 to B8. According to the multiple groups of adjustment parameters obtained, the optimization degree corresponding to each group of adjustment parameters is calculated, and the preset parameters corresponding to the minimum optimization degree are used as the preset parameters corresponding to the group of abnormal prediction data. The preset parameters corresponding to v groups of abnormal prediction data are obtained by analogy, and the mean values ​​of multiple preset parameters (i.e., the initialization temperature mean, the minimum temperature mean, and the maximum number of iterations mean) are used as the initialization temperature preset in step B2. , minimum temperature And the maximum number of iterations .

[0110] In the above step B3, the method for constructing M groups of adjustment parameters is: obtaining the shore power current in the abnormal prediction data; dividing the shore power current into N values, the sum of the N values ​​is the shore power current, and N is the number of ships connected to the shore power equipment; taking the N values ​​as a group of adjustment parameters, and constructing a total of M groups of adjustment parameters, the M groups of adjustment parameters are not consistent, and the values ​​in the adjustment parameters correspond to the ships one by one.

[0111] In the above step B4, the expression of the optimization function is: ;

[0112] In the formula, To optimize, is the reverse electrical probability;

[0113] The method to obtain the reverse electrical probability is: The corresponding adjustment parameters and abnormal prediction data are used as research data. The research data are input into the trained probability prediction model to predict the corresponding reverse power probability. The specific training process of the probability prediction model is consistent with the training process of the shore power voltage prediction model, and both are deep neural network models.

[0114] Determine the temperature drop factor The steps include:

[0115] Step C1: Preset initial coefficient search interval , this embodiment preferably , ; Set the split ratio , this embodiment preferably ;

[0116] Step C2: According to the segmentation ratio , the initial coefficient search interval Divide two new points and , , ; Construct an optimization set, which includes different cooling coefficients And the corresponding optimization degree;

[0117] Step C3: Obtain according to the optimization degree set and The corresponding optimization degree will be The corresponding optimization degree is marked as the first optimization degree. The corresponding optimization degree is marked as the second optimization degree; the method for constructing the optimization degree set is: in the process of formulating the historical dynamic adjustment strategy, the technicians in this field collect v groups of different abnormal prediction data, and pre-set Y different cooling coefficients for a group of abnormal prediction data in turn. , according to step B2 to step B8, the adjustment parameters are obtained, and the optimization degree corresponding to each set of adjustment parameters is calculated according to the obtained multiple sets of adjustment parameters as the corresponding temperature reduction coefficient The optimization degree of each cooling coefficient is obtained by analogy. The corresponding v optimization degrees are each cooling coefficient The corresponding v mean values ​​of the optimization degree are used as the cooling coefficient for each Corresponding optimization degree; according to Y different cooling coefficients Corresponding optimization degree, constructing optimization degree set;

[0118] Step C4: If the first optimization degree is greater than or equal to the second optimization degree, the minimum value of the updated coefficient search interval is the point corresponding to the first optimization degree, and the maximum value is the maximum value of the coefficient search interval before updating; if the first optimization degree is less than the second optimization degree, the minimum value of the updated coefficient search interval is the minimum value of the coefficient search interval before updating, and the maximum value is the point corresponding to the second optimization degree; illustratively, the initial coefficient search interval Passing point and After updating, if the first optimization degree is greater than or equal to the second optimization degree, the update coefficient search interval is , if the first optimization degree is less than the second optimization degree, the update coefficient search interval is ;

[0119] Step C5: preset a width threshold, which is preset by those skilled in the art according to actual conditions, and obtain the interval width by subtracting the minimum value from the maximum value in the updated coefficient search interval; compare the interval width with the width threshold; if the interval width is greater than or equal to the width threshold, the updated coefficient search area is divided according to the segmentation ratio. , re-dividing two new points and , and return to step C3; if the interval width is less than the width threshold, the average of the maximum and minimum values ​​in the updated coefficient search area is used as the cooling coefficient .

[0120] The reverse current warning module is used to execute dynamic adjustment strategies and determine whether to generate a warning instruction.

[0121] When the time corresponding to the abnormal prediction data is reached, the dynamic adjustment strategy is executed.

[0122] Methods for determining whether to generate an early warning instruction include:

[0123] If the reverse power type is reverse flow between ships, the current flow direction of the abnormal ship is collected through the Hall effect sensor installed at the power input end of the abnormal ship, and marked as the ship flow direction value; if the ship flow direction value is a positive value, no warning instruction is generated, indicating that the current flow direction of the abnormal ship is consistent with the normal direction; if the ship flow direction value is a negative value, a warning instruction is generated, indicating that the current flow direction of the abnormal ship is inconsistent with the normal direction; if the reverse power type is shore power reverse flow, the current flow direction of the shore power equipment is collected through the Hall effect sensor installed at the output end of the shore power equipment, and marked as the shore current direction value; if the shore current direction value is a positive value, no warning instruction is generated, indicating that the current flow direction of the shore power equipment is consistent with the normal direction; if the shore current direction value is a negative value, a warning instruction is generated, indicating that the current flow direction of the shore power equipment is inconsistent with the normal direction.

[0124] This embodiment collects historical power data and uses deep learning technology to predict future power data, accurately detects reverse current phenomena and identifies reverse current types, so as to formulate targeted dynamic adjustment strategies; and after the dynamic adjustment strategy is executed, the execution of the adjustment strategy is judged in real time, and reverse current warning is carried out in time; through the collaborative work of modules such as data collection, predictive analysis, reverse current detection and dynamic adjustment, it can effectively monitor and prevent current reverse flow phenomena and improve the safety and stability of the shore power system; it not only optimizes power supply management and reduces the risk of equipment damage, but also improves the overall operating efficiency and safety and reliability of the power system.

[0125] Example 2

[0126] The present application also provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories store computer-readable codes, and when the computer-readable codes are run by the one or more processors, they may execute a connected power supply anti-reverse power intelligent control system as described above.

[0127] The method or system according to the implementation mode of the present application can also be implemented with the help of the architecture of the electronic device shown in the present application. The electronic device may include a bus, one or more CPUs, ROM, RAM, a communication port connected to a network, input / output, a hard disk, etc. A storage device in the electronic device, such as a ROM or a hard disk, can store a connected power supply and anti-reverse power intelligent control system provided in the present application. Furthermore, the electronic device may also include a user interface. Of course, the architecture shown in the present application is only exemplary. When implementing different devices, one or more components in the electronic device shown in the present application may be omitted according to actual needs.

[0128] Example 3

[0129] One embodiment of the present application discloses a computer-readable storage medium. Computer-readable instructions are stored on the computer-readable storage medium. When the computer-readable instructions are executed by the processor, a connection power supply anti-reverse power intelligent control system according to the embodiment of the present application described with reference to the above figures can be executed. The storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0130] In addition, according to the implementation of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, which stores machine-readable instructions, and the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided by the present application, for example: a connected power supply anti-reverse power intelligent control system. When the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0131] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0132] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent control system for connecting power supply and preventing reverse electricity, characterized in that: include: Data acquisition module, used to collect historical power data; Data prediction module, used to analyze historical power data and predict future power data; The reverse power detection module is used to analyze future power data and determine whether to generate a reverse power instruction; The method for determining whether to generate a reverse power instruction comprises: According to the future power data, calculate the current change and load change at the next m moments; use the current change and load change at the next m moments and the shore power current in the future power data as test data; build a detection forest according to the test data; calculate the actual distance and expected distance of each data in the future data according to the detection forest; calculate the abnormal value of each data in the future data according to the actual distance and expected distance of each data in the future data; preset an abnormal threshold; if the abnormal value of the data in the future data is greater than or equal to the abnormal threshold, generate a reverse power instruction; if the abnormal value of each data in the future data is less than the abnormal threshold, do not generate a reverse power instruction; A reverse power analysis module is used to analyze the reverse power type based on future power data if a reverse power instruction is generated; The dynamic adjustment module is used to integrate the reverse power type and future power data to formulate a dynamic adjustment strategy; the steps of formulating the dynamic adjustment strategy include: Step B1: If the reverse power type is inter-ship reverse power, proceed to step B2; if the reverse power type is shore power reverse power, proceed to step B9; Step B2: Preset initialization temperature T max , minimum temperature T min And the maximum number of iterations ξ, and let the current temperature T = T max ; Step B3: construct M groups of adjustment parameters, where M is an integer greater than 1; randomly set a feasible solution χ, where the feasible solution χ is the adjustment parameter, and the range of the feasible solution χ is the M groups of parameters; Step B4: Determine the optimization function and the temperature reduction coefficient δ; Step B5: Calculate the optimization degree f corresponding to the feasible solution χ; take the feasible solution χ as the current point, perform random perturbations in the neighborhood of the current point, obtain a new feasible solution χ′, and calculate the optimization degree f′ corresponding to the new feasible solution χ′; Step B6: Calculate the optimization difference f″, the expression of the optimization difference f″ is f″=ff′; if the optimization difference f″>0, set χ=χ′; if the optimization difference f″≤0, calculate the probability p′, and set χ=χ′ according to the probability p′; the expression of the probability p′ is: In the formula, e is a natural constant; Step B7: looping steps B5 to B6 until the number of loops reaches the maximum number of iterations ξ, the loop ends; let the current temperature T = T × δ; let the maximum number of iterations ξ = ξ × δ; Step B8: Repeat steps B5 to B7 until the current temperature T < T min When , the loop ends and the adjustment parameters corresponding to the feasible solution χ are obtained; Step B9: Subtract the shore power voltage from the ship voltage of the abnormal ship, and then multiply by k to obtain a voltage adjustment value, k∈(1,1.1); The reverse current warning module is used to execute dynamic adjustment strategies and determine whether to generate a warning instruction.

2. According to claim 1, a connection power supply anti-reverse power intelligent control system is characterized in that: The historical power data is power data at historical moments, and the historical power data includes n groups of power data, where n is an integer greater than 1, and one group of power data corresponds to one moment; the power data includes shore power data and ship data; the shore power data includes shore power voltage and shore power current; the ship data includes ship voltage, ship current and ship load, and the ship data corresponds one-to-one to the ship connected to the shore power equipment; The method for predicting future power data comprises: Collect real-time and future environmental data; The real-time environmental data is the environmental data at the current moment, and the future environmental data is the environmental data at the next m-1 moments, where m is an integer greater than 2. The environmental data includes environmental temperature, environmental humidity and extreme weather data; different digital labels are set for different extreme weather data and marked as weather labels; the extreme weather data in the environmental data is replaced with the corresponding weather labels, and the historical power data and real-time environmental data are used as analysis data, and the analysis data is input into the trained power forecasting model to predict the power data at the next moment and marked as prediction data; the earliest power data at the corresponding moment in the historical power data is replaced with prediction data, and the historical power data is marked as replacement power data; the historical power data in the analysis data is replaced with replacement power data, and the real-time environmental data is replaced with the environmental data in the future environmental data that is the same as the environmental data at the corresponding moment of the last predicted prediction data, and is input into the trained power forecasting model again to predict the power data at the next moment of the corresponding moment of the last predicted prediction data; in this cycle, a total of m groups of prediction data are predicted, and the m groups of prediction data correspond to m moments in the future, and the m groups of prediction data are used as future power data.

3. According to claim 2, a connection power supply anti-reverse power intelligent control system is characterized in that: The power prediction model includes a shore power voltage prediction model, a shore power current prediction model, a ship voltage prediction model, a ship current prediction model and a ship load prediction model; the power prediction model corresponds to the power data one by one, and the training process of each model in the power prediction model is consistent, and they are all deep neural network models; The training process of the shore power voltage prediction model includes: P groups of analysis data are collected in advance, corresponding shore power voltages are set for the p groups of analysis data, where p is an integer greater than 1, and the analysis data and the corresponding shore power voltage are converted into a corresponding set of feature vectors; each set of feature vectors is used as the input of a shore power voltage prediction model, and the shore power voltage prediction model uses a set of predicted shore power voltages corresponding to each set of analysis data as output, and uses the actual shore power voltage corresponding to each set of analysis data as a prediction target, where the actual shore power voltage is the pre-set shore power voltage corresponding to the analysis data; minimizing the sum of prediction errors of all analysis data is used as a training target; and the shore power voltage prediction model is trained until the sum of prediction errors converges and the training is stopped.

4. According to claim 3, the intelligent control system for connecting power supply and preventing reverse electricity is characterized in that: The method for calculating the current change amount and the load change amount at the next m moments includes: The latest power data at the corresponding moment in the historical power data is obtained and marked as real-time power data; the ship current in the real-time power data and the ship current in the future power data are sorted from late to early according to the corresponding moment; each ship current is subtracted from the ship current ranked next in positive order to obtain the current change at the next m moments; the ship load in the real-time power data and the ship load in the future power data are sorted from late to early according to the corresponding moment; each ship load is subtracted from the ship load ranked next in positive order to obtain the load change at the next m moments.

5. According to claim 4, a connection power supply anti-reverse power intelligent control system is characterized in that: The steps of constructing the detection forest include: Step A1: according to the test data, obtaining a data range value, the data range value includes a current variation range value, a load variation range value and a shore power current range value; Step A2: Randomly select a type of data from the test data and mark it as selected data; Step A3: randomly select a value in the data range corresponding to the selected data as a split point, and split the selected data according to the split point to obtain a split sub-set; Step A4: Obtain a numerical range for each slicing sub-set, wherein the maximum value of the numerical range is the maximum value of the numerical range in the slicing sub-set, and the minimum value of the numerical range is the minimum value of the numerical range in the slicing sub-set; randomly select a slicing point in the numerical range corresponding to each slicing sub-set, and obtain a total of q slicing points, where q is the number of slicing sub-sets; and slicing each slicing sub-set according to the q slicing points; Step A5: loop step A4 until there is only one value in each segmentation sub-set, the loop ends, obtains the detection tree corresponding to the selected data, and proceeds to step A6; Step A6: looping steps A2 to A5 until each type of data in the test data is marked as selected data, the loop ends, and the detection trees corresponding to each selected data are combined into a detection forest; Methods for calculating actual distance and expected distance include: The actual distance is calculated as follows: the number of times the data is split in each corresponding detection tree is counted and marked as the number of splits. Each number of splits is added up in sequence and then divided by the number of detection trees corresponding to the data to obtain the actual distance corresponding to the data. The expected distance is calculated as: In the formula, qw is the expected distance; Methods for calculating outliers include: Where yc is the outlier and sj is the actual distance.

6. The intelligent control system for connecting power supply and preventing reverse electricity according to claim 5, characterized in that: The types of reverse power include inter-ship reverse power and shore power reverse power; The method for analyzing the reverse electricity type includes: Obtain data in the future data corresponding to the reverse power instruction and mark it as abnormal data; obtain the predicted data corresponding to the moment corresponding to the abnormal data and mark it as abnormal predicted data; obtain the shore power voltage and ship voltage in the abnormal predicted data; compare the shore power voltage with each ship voltage respectively; if the shore power voltage is greater than or equal to each ship voltage, analyze the reverse power type as inter-ship reverse power; if there is a ship voltage greater than the shore power voltage, analyze the reverse power type as shore power reverse power, and mark the ship corresponding to the ship voltage as an abnormal ship.

7. The intelligent control system for connecting power supply and preventing reverse electricity according to claim 6, characterized in that: In step B3, the method for constructing M groups of adjustment parameters is: obtaining the shore power current in the abnormal prediction data; dividing the shore power current into N values, the sum of the N values ​​is the shore power current, and N is the number of ships connected to the shore power equipment; taking the N values ​​as a group of adjustment parameters, constructing a total of M groups of adjustment parameters, the M groups of adjustment parameters are all inconsistent, and the values ​​in the adjustment parameters correspond to the ships one by one; In step B4, the expression of the optimization function is: f=ng; In the formula, f is the optimization degree, ng is the reverse electric probability; The method for obtaining the reverse electricity probability is: take the adjustment parameters corresponding to the feasible solution χ and the abnormal prediction data as research data, input the research data into the trained probability prediction model, and predict the corresponding reverse electricity probability; the training process of the probability prediction model is consistent with the training process of the shore power voltage prediction model, and both are deep neural network models.

8. The intelligent control system for connecting power supply and preventing reverse electricity according to claim 7, characterized in that: The steps for determining the temperature drop coefficient δ include: Step C1: preset the initial coefficient search interval [a0, b0]; set the segmentation ratio h; Step C2: According to the segmentation ratio h, the initial coefficient search interval [a0, b0] is divided into two new points ζ1 and ζ2, ζ1 = a0 + (1-h)(b0-a0), ζ2 = a0 + h(b0-a0); construct an optimization degree set, which includes different cooling coefficients δ and corresponding optimization degrees; Step C3: Obtain the optimization degrees corresponding to ζ1 and ζ2 respectively according to the optimization degree set, mark the optimization degree corresponding to ζ1 as the first optimization degree, and mark the optimization degree corresponding to ζ2 as the second optimization degree; Step C4: if the first optimization degree is greater than or equal to the second optimization degree, the minimum value of the updated coefficient search interval is the point corresponding to the first optimization degree, and the maximum value is the maximum value of the coefficient search interval before updating; if the first optimization degree is less than the second optimization degree, the minimum value of the updated coefficient search interval is the minimum value of the coefficient search interval before updating, and the maximum value is the point corresponding to the second optimization degree; Step C5: preset a width threshold, subtract the minimum value from the maximum value in the updated coefficient search interval to obtain the interval width; compare the interval width with the width threshold; if the interval width is greater than or equal to the width threshold, then re-divide the updated coefficient search area into two new points ζ1 and ζ2 according to the segmentation ratio h, and return to step C3; if the interval width is less than the width threshold, then take the average of the maximum and minimum values ​​in the updated coefficient search area as the cooling coefficient δ.

9. The intelligent control system for connecting power supply and preventing reverse power supply according to claim 8, characterized in that: When the time corresponding to the abnormal prediction data is reached, the dynamic adjustment strategy is executed; The method for determining whether to generate an early warning instruction includes: If the reverse power type is reverse flow between ships, the current flow direction of the abnormal ship is collected and marked as the ship flow direction value; if the ship flow direction value is a positive value, no warning instruction is generated; if the ship flow direction value is a negative value, a warning instruction is generated; if the reverse power type is shore power reverse flow, the current flow direction of the shore power equipment is collected and marked as the shore current direction value; if the shore current direction value is a positive value, no warning instruction is generated; if the shore current direction value is a negative value, a warning instruction is generated.

Images