Energy-saving management method based on AI algorithm and multiple collection methods

Abstract

The present invention discloses an energy-saving management method based on AI algorithm and multiple collection methods, belonging to the field of energy management technology. It aims at the problem that the static mode of existing household appliance energy monitoring is difficult to capture subtle changes, and the potential fault detection is delayed, resulting in energy waste. The collection frequency is dynamically adjusted according to the power used by the household appliances, and the changes in the operating status of the appliances can be captured more accurately in combination with the power jump duration and number. This makes the monitoring of potential faults of household appliances more sensitive, and subtle anomalies can be discovered in time, thereby giving timely warnings. The health detection parameters are analyzed to obtain wave evaluation values, change evaluation values, and trend evaluation values, which are integrated into health evaluation values, and then judged according to the preset interval. When the appliance is in a slightly abnormal state, a warning instruction can be generated in time and sent to the user's mobile phone terminal, so that the user can perform maintenance and inspection on the appliance in advance, effectively prevent the occurrence of faults, and avoid energy waste caused by faults.

Description

Technical Field

[0001] The present invention belongs to the technical field of energy management, and specifically relates to an energy-saving management method based on AI algorithm and multiple collection methods. Background Art

[0002] With the rapid development of science and technology and the continuous improvement of people's living standards, smart home products have gradually entered thousands of households, and the intelligence level of household appliances has been increasing. At the same time, the world's attention to energy conservation and sustainable development is also increasing. As an important unit of energy consumption, how to achieve efficient energy-saving management of household appliances has become an important issue that needs to be solved urgently.

[0003] The existing household appliance energy monitoring method adopts a static monitoring mode, that is, limited data collection of household appliances at fixed time intervals or under specific conditions, so that it is difficult to capture subtle changes in appliances during long-term operation. This often leads to a delay in the discovery of potential faults in household appliances, and failures cannot be prevented in time, which in turn leads to energy waste caused by faults. For this reason, we propose an energy-saving management method based on AI algorithms and multiple collection methods. Summary of the invention

[0004] The purpose of the present invention is to provide an energy-saving management method based on AI algorithm and multiple collection methods to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: an energy-saving management method based on AI algorithm and multiple collection methods, comprising the following steps:

[0006] Step 1: Match the initial collection frequency according to the power of the household appliance when it is in use, and switch the collection frequency according to the power jump duration and jump times when the household appliance is running. The household appliance collects the health detection parameters of the household appliance itself according to the collection frequency corresponding to each moment;

[0007] Step 2: Establish a connection with the cloud AI service platform to receive the health detection parameters of the home appliances, analyze the received health detection parameters, obtain the wave evaluation value, change evaluation value and trend evaluation value, and obtain the health evaluation value after comprehensive analysis of the three. Monitor the operating status of the home appliances according to the health evaluation value. If the monitoring result shows that maintenance is needed, generate a matching maintenance instruction containing the name of the home appliance;

[0008] Step 3: The server is used to manage user information and maintenance personnel information;

[0009] Step 4: When a matching maintenance instruction containing the name of a household appliance is received, the matching value of the maintenance personnel is calculated based on the name of the household appliance to be repaired and the corresponding user information and maintenance personnel information, and the selected maintenance personnel are selected to repair the household appliance based on the matching value of the maintenance personnel.

[0010] Preferably, in step 1, the specific process of matching the initial acquisition frequency according to the power of the household appliance when in use is:

[0011] By connecting household appliances to smart sockets, the real-time power of household appliances when in use can be obtained. Household appliances include: refrigerators, televisions, washing machines, air conditioners, electric water heaters, and electric ovens. For each household appliance, multiple power collection intervals are preset, and each power collection interval corresponds to a basic collection frequency. At the same time, the power collection duration is preset. The average power of the household appliance during the power collection duration is recorded with the startup time of the household appliance as the starting time. The average power of the household appliance during the power collection duration is matched with all power collection intervals, and the corresponding basic collection frequency is output as the initial collection frequency when the household appliance is started.

[0012] Preferably, in step 1, the acquisition frequency is switched according to the power jump duration and the number of jumps, and the household appliance collects the health detection parameters of the household appliance itself according to the acquisition frequency corresponding to each moment. The specific process is:

[0013] After the startup time of the household appliance is longer than the preset power collection time, if the power value of the household appliance jumps from the original power collection interval to another power collection interval at a certain moment, then this moment is marked as the power jump moment, and a time jump threshold is preset. Starting from the power jump moment, the duration of the power value of the household appliance jumping from the original power collection interval to another power collection interval is detected, which is recorded as the jump duration. If the jump duration is greater than or equal to the preset time jump threshold, the basic collection frequency of the household appliance is updated to the basic collection frequency corresponding to the current power collection interval;

[0014] If the jump duration is less than the preset time jump threshold, starting from the power jump moment, the number of times the power value of the household appliance jumps across the power collection interval is detected, recorded as the number of jumps, and the jump number threshold is preset. If the jump number is greater than or equal to the preset jump number threshold, the emergency collection frequency is preset, and the original initial collection frequency of the household appliance is increased to the emergency collection frequency; if the jump number is less than the preset jump number threshold, the collection frequency of the household appliance remains unchanged;

[0015] When the emergency collection frequency is increased, if the power value of the household appliance remains within a certain power collection interval and the duration is greater than or equal to the preset time jump threshold, the collection frequency of the household appliance is gradually restored from the emergency collection frequency to the basic collection frequency corresponding to the power collection interval;

[0016] The household appliances collect health detection parameters of the household appliances themselves according to the corresponding collection frequency at each moment; the health detection parameters include: power value, surface temperature value, vibration frequency, and noise value.

[0017] Preferably, in step 2, the received health detection parameters are analyzed to obtain the specific process of wave evaluation value:

[0018] For each household appliance, the operation time of the household appliance is divided into several continuous and non-overlapping detection time periods at fixed intervals;

[0019] For each detection period, the standard deviation of each health detection parameter in the detection period is calculated to obtain the health fluctuation value, which includes: power fluctuation value, temperature fluctuation value, vibration fluctuation value and noise fluctuation value;

[0020] It is preset that each health fluctuation value corresponds to a fluctuation threshold. The health fluctuation value is compared with the corresponding fluctuation threshold. If the health fluctuation value is greater than the corresponding fluctuation threshold, the difference between the health fluctuation value and the corresponding fluctuation threshold is calculated and marked as the health wave difference JKi, where i is the number of the health wave difference, i=, 1, 2...n, n≤4; n is the total number of health wave difference numbers, using the formula: , and obtain the wave evaluation value PZ, where ai is the preset weight coefficient, and ai includes: a1, a2, a3, and a4.

[0021] Preferably, in step 2, the received health detection parameters are analyzed to obtain the specific process of the variable evaluation value as follows:

[0022] For each health detection parameter, let the value of the health detection parameter at different collection points during the detection period be P g , g is the number of the collection point, g = 1, 2...m; m is the total number of collection points, using the formula: , get the change rate value PH of the health detection parameter, where T g is the collection time; each health detection parameter is preset to have a corresponding change rate threshold. If the change rate value of the health detection parameter is greater than the corresponding threshold, the deviation value between the change rate value of the health detection parameter and the corresponding threshold is calculated and marked as the health deviation value JPs, where s is the number of the health deviation value, s=, 1, 2...e, e≤4, and e is the total number of health deviation value numbers; using the formula: , and obtain the variable evaluation value BZ, where bi is the preset weight coefficient, bi includes: b1, b2, b3, b4.

[0023] Preferably, in step 2, the specific process of analyzing the received health detection parameters to obtain the trend evaluation value is:

[0024] For each health detection parameter in each detection period, collect the historical data of household appliances in the detection period, use the historical data of the health detection parameters in the detection period to fit the linear regression line equation of the health detection parameters, and for each collection point, calculate its corresponding value on the linear regression line equation of the health detection parameters and mark it as the trend value Y g , the actual value P g and trend value Y g Perform the difference and get the real bias value △P g , using the formula: , get the trend deviation value QD;

[0025] Obtain the trend deviation value QDr corresponding to each health detection parameter within the detection period, where r is the number of the trend deviation value corresponding to each health detection parameter, r=1, 2...h; h is the total number of trend deviation value numbers; use the formula: , and obtain the trend evaluation value QP, where cr is the preset weight coefficient, cr includes: c1, c2, c3, c4.

[0026] Preferably, in step 2, the specific process of obtaining the health evaluation value and monitoring the operating status of the household appliance according to the health evaluation value is:

[0027] By normalizing the wave assessment value PZ, the change assessment value BZ, and the trend assessment value QP, and substituting them into the preset formula: JPZ=PZ×d1+BZ×d2+QP×d3, the health assessment value JPZ is obtained, where d1, d2, and d3 are preset weight coefficients;

[0028] For each household appliance, three health assessment value intervals are preset, namely, the health assessment value safety interval, the health assessment value warning interval, and the health assessment value maintenance interval, and the health assessment value corresponding to the detection period is matched with the three health assessment value intervals;

[0029] If the health assessment value is matched to the safety range, no processing is required;

[0030] If the health assessment value warning interval is matched, a warning instruction containing the name of the household appliance is generated and sent to the user's mobile terminal. After receiving the warning instruction, the user performs maintenance and inspection on the household appliance corresponding to the warning instruction;

[0031] If the health assessment value matches the maintenance range, the household appliances are shut down and stopped, and a maintenance instruction containing the name of the household appliances is generated and sent to the user's mobile terminal. When the user receives the maintenance instruction for the corresponding appliance, he or she can reply whether to perform the maintenance. If the user replies to perform the maintenance, a matching maintenance instruction containing the name of the household appliance is generated.

[0032] Preferably, the working process of step three is:

[0033] User information management includes: storage of user registration information, including name, phone number and home address;

[0034] Maintenance personnel information management: used to register maintenance personnel's maintenance information, including name, phone number, and length of service; at the same time, a location positioning connection is established with the maintenance personnel's mobile phone. In addition, it is also responsible for recording the maintenance personnel's maintenance points, which are reset once a month.

[0035] Preferably, the working process of step 4 is:

[0036] When a matching maintenance instruction containing the name of a household appliance is received, the name of the household appliance to be repaired is extracted; the household appliance to be repaired is marked as a device to be repaired, and the user submits an appointment time for the device to be repaired through a mobile terminal;

[0037] Obtain the corresponding maintenance personnel from the server according to the name of the equipment to be repaired, and mark them as preliminary candidates;

[0038] Screen the preliminary candidates who do not have a maintenance appointment on the appointment date and mark them as candidates Rx, where x=1, ..., p; p is the total number of candidates;

[0039] Obtain the user's address and the address of the candidate from the server and calculate the distance difference to obtain the maintenance distance FRx; extract the candidate's maintenance service years SRx and the candidate's maintenance points JR from the server

[0040] After normalizing the maintenance distance FRx, maintenance service years SRx, and maintenance points JRx, the matching value WRx is obtained using the formula: WRx=FRx×e1+SRx×e2+JRx×e3, where e1, e2, and d3 are preset proportional coefficients;

[0041] The candidate with the largest matching value is marked as the selected maintenance personnel of the user, and a message about the maintenance and the maintenance appointment time is sent to the selected maintenance personnel's mobile terminal; at the same time, the name and contact number of the selected maintenance personnel are sent to the user's mobile terminal. After the selected maintenance personnel completes the maintenance, the maintenance completion information is sent to the server.

[0042] Preferably, the processing process after the server receives the maintenance completion information is:

[0043] After receiving the repair completion instruction sent by the selected maintenance personnel, the server generates maintenance score evaluation information and sends it to the user's mobile phone terminal where the selected maintenance personnel last accepted the repair. The score evaluation information includes: whether the door-to-door service is on time (1 to 5 points), the maintenance quality (1 to 5 points) and the service attitude (1 to 5 points);

[0044] After receiving the maintenance score evaluation information, the user scores the maintenance service provided by the selected maintenance personnel, and the total score scored by the user is added to the maintenance score stored in the server by the selected maintenance personnel.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] (1) This energy-saving management method based on AI algorithm and multiple collection methods can dynamically adjust the collection frequency according to the power used by household appliances through multiple collection methods, and can more accurately capture the changes in the operating status of appliances by combining factors such as the duration and number of power jumps. This makes the monitoring of potential faults of household appliances more sensitive and can detect subtle abnormalities in a timely manner. For example, when the power collection interval jumps frequently in a short period of time, the collection frequency is increased, which helps to accurately capture the details of potential problems such as intermittent motor overload and poor circuit contact, thereby providing timely warnings. The health detection parameters are analyzed to obtain wave evaluation values, change evaluation values, and trend evaluation values, which are integrated into health evaluation values. Then, according to the preset interval, a judgment can be made. When the appliance is in a slightly abnormal state, a warning instruction can be generated in time and sent to the user's mobile terminal, so that the user can perform maintenance and inspection on the appliance in advance, effectively prevent the occurrence of faults, avoid energy waste caused by faults, and achieve energy-saving management.

[0047] (2) This energy-saving management method based on AI algorithm and multiple data collection methods selects maintenance personnel by calculating the matching value based on multiple factors such as the name of the household appliance, maintenance appointment time, maintenance address, and the maintenance personnel's address, length of service, and points. This method comprehensively considers factors such as distance, experience, and service quality, and can accurately match users with more suitable maintenance personnel, improve maintenance efficiency and quality, and reduce the increase in energy consumption caused by untimely or improper maintenance. Maintenance personnel with long maintenance experience, high points, and close proximity to users are more likely to be selected, thereby quickly solving appliance problems and restoring the appliance to normal operation, thereby achieving energy-saving purposes.

[0048] (3) This energy-saving management method based on AI algorithm and multiple data collection methods resets the maintenance personnel's points once a month, and after the maintenance, the user can score the maintenance personnel's punctuality, maintenance quality and service attitude, and the scoring results are added to the maintenance personnel's points; this encourages maintenance personnel to improve service quality, visit on time, ensure maintenance quality and provide good service attitude, thereby improving the efficiency and quality of the entire maintenance service system, and indirectly contributing to energy-saving management.

[0049] (4) This energy-saving management method based on AI algorithms and multiple collection methods enables users to receive early warning information and maintenance information of electrical appliances in a timely manner, making it convenient to arrange electrical appliance maintenance and repair matters, thereby improving users' control over the operating status of household appliances and the convenience of use; at the same time, users participate in the service evaluation of maintenance personnel, which enhances the interaction between users and the maintenance service system; this interaction helps to improve users' understanding of energy-saving management, allowing users to pay more attention to the maintenance and rational use of electrical appliances in their daily use, thereby further promoting household energy conservation. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a structural schematic diagram of the present invention. DETAILED DESCRIPTION

[0051] 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.

[0052] Embodiment 1

[0053] See also Figure 1 , the present invention provides an energy-saving management method based on AI algorithm and multiple collection methods, a data collection module, an AI algorithm analysis module, an intelligent matching module, and a server;

[0054] The data collection module uses the power of household appliances as a factor, presets the power collection interval and the corresponding collection frequency, and dynamically adjusts the collection frequency according to the startup time and power jump situation, including determining the initial frequency at startup, flexibly switching the collection frequency according to the power jump time and the number of jumps during operation, and an emergency frequency recovery mechanism. The health and energy-saving parameters of household appliances are collected according to the corresponding collection frequency of the household appliances themselves, and sent to the AI ​​algorithm analysis module respectively. The specific process is as follows:

[0055] By connecting household appliances to smart sockets, the real-time power of household appliances when in use is obtained. Household appliances include but are not limited to: refrigerators, televisions, washing machines, air conditioners, electric water heaters, and electric ovens; for each household appliance, multiple power collection intervals are preset, and each power collection interval corresponds to a basic collection frequency. At the same time, the power collection duration is preset, and the average power of the household appliance during the power collection duration is recorded with the start time of the household appliance as the starting time. The average power of the household appliance during the power collection duration is matched with all power collection intervals, and the corresponding basic collection frequency is output as the initial collection frequency when the household appliance is started;

[0056] After the startup time of the household appliance is longer than the preset power collection time, if the power value of the household appliance jumps from the original power collection interval to another power collection interval at a certain moment, then this moment is marked as the power jump moment, and a time jump threshold is preset. Starting from the power jump moment, the duration of the power value of the household appliance jumping from the original power collection interval to another power collection interval is detected, which is recorded as the jump duration. If the jump duration is greater than or equal to the preset time jump threshold, the basic collection frequency of the household appliance is updated to the basic collection frequency corresponding to the current power collection interval;

[0057] If the jump duration is less than the preset time jump threshold, starting from the power jump moment, the number of times the power value of the household appliance jumps across the power collection interval is detected, recorded as the number of jumps, and the jump number threshold is preset. If the jump number is greater than or equal to the preset jump number threshold, the emergency collection frequency is preset, and the original initial collection frequency of the household appliance is increased to the emergency collection frequency, where the emergency collection frequency is much higher than the original basic collection frequency; It should be noted that frequent power collection interval jumps in a short period of time may indicate that there are potential unstable factors or hidden faults in the electrical equipment, such as intermittent overload of the motor, poor circuit contact and other problems. Increasing the collection frequency helps to more accurately capture the details of these abnormal changes, so as to timely discover the root cause of the problem and take corresponding measures to avoid further deterioration of the fault and ensure the safe and stable operation of the electrical equipment; If the number of jumps is less than the preset jump number threshold, the collection frequency of the household appliance remains unchanged; It should be noted that: This indicates that although the power collection interval jump occurs, the number of jumps does not reach the threshold in a short period of time, which may be normal equipment operation fluctuations or accidental external interference factors, such as the influence of instantaneous fluctuations in grid voltage. There is no need to excessively adjust the collection frequency to avoid increasing system burden and data storage pressure due to unnecessary high-frequency collection;

[0058] When the emergency collection frequency is increased, if the power value of the household appliance remains within a certain power collection interval and the duration is greater than or equal to the preset time jump threshold, the collection frequency of the household appliance is gradually restored from the emergency collection frequency to the basic collection frequency corresponding to the power collection interval; in this way, while ensuring that potential problems are discovered and handled in a timely manner, system resources can be reasonably utilized to avoid excessive consumption of system performance and resources caused by being in a high-frequency collection state for a long time;

[0059] Household appliances collect health detection parameters of themselves according to the corresponding collection frequency at each moment, and send the health detection parameters collected in the same batch to the AI ​​algorithm analysis module; the health detection parameters include: power value, surface temperature value, vibration frequency, and noise value.

[0060] The AI ​​algorithm analysis module establishes a connection with the cloud AI service platform through a high-speed network interface, receives the health detection parameters of household appliances, and transmits the received health detection parameters to the cloud AI service platform. The cloud AI service platform analyzes the received data, obtains the health evaluation value, analyzes the health evaluation value, and monitors the operating status of household appliances based on the health evaluation value. The specific process is as follows:

[0061] For each household appliance, the operation time of the household appliance is divided into several continuous and non-overlapping detection time periods at fixed intervals;

[0062] For each detection period, the standard deviation of each health detection parameter in the detection period is calculated to obtain the health fluctuation value, which includes: power fluctuation value, temperature fluctuation value, vibration fluctuation value and noise fluctuation value;

[0063] It is preset that each health fluctuation value corresponds to a fluctuation threshold. The health fluctuation value is compared with the corresponding fluctuation threshold. If the health fluctuation value is greater than the corresponding fluctuation threshold, the difference between the health fluctuation value and the corresponding fluctuation threshold is calculated and marked as the health wave difference JKi, where i is the number of the health wave difference, i=, 1, 2...n, n≤4; n is the total number of health wave difference numbers, using the formula: , get the wave evaluation value PZ, where ai is the preset weight coefficient, ai includes: a1, a2, a3, a4; the larger the wave evaluation value, the worse the stability of the household appliance during operation;

[0064] For each health detection parameter, let the value of the health detection parameter at different collection points during the detection period be P g , g is the number of the collection point, g = 1, 2...m; m is the total number of collection points, using the formula: , get the change rate value PH of the health detection parameter, where T gis the collection time; each health detection parameter is preset to have a corresponding change rate threshold. If the change rate value of the health detection parameter is greater than the corresponding threshold, the deviation value between the change rate value of the health detection parameter and the corresponding threshold is calculated and marked as the health deviation value JPs, where s is the number of the health deviation value, s=, 1, 2...e, e≤4, and e is the total number of health deviation value numbers; using the formula: , get the variable evaluation value BZ, where bi is the preset weight coefficient, bi includes: b1, b2, b3, b4; the larger the variable evaluation value, the more drastic the change of the health detection parameters of the appliance during operation, and the greater the possibility of failure risk;

[0065] For each health detection parameter in each detection period, collect the historical data of household appliances in the detection period, and use the historical data of the health detection parameters in the detection period to fit the linear regression line equation of the health detection parameters, where the equation can be: Y=kt+b, t is time, Y is the health detection parameter value, k is the slope, and b is the intercept. For each collection point, calculate its corresponding value on the linear regression line equation of the health detection parameter and mark it as the trend value Y g , the actual value P g and trend value Y g Perform the difference and get the real bias value △P g , using the formula: , get the trend deviation value QD;

[0066] Obtain the trend deviation value QDr corresponding to each health detection parameter within the detection period, where r is the number of the trend deviation value corresponding to each health detection parameter, r=1, 2...h; h is the total number of trend deviation value numbers; use the formula: , get the trend evaluation value QP, where cr is the preset weight coefficient, cr includes: c1, c2, c3, c4; the larger the trend evaluation value, the more serious the deviation of the actual value of the health detection parameter from the normal trend during the detection period, indicating that the possibility of failure of the household appliance is greater;

[0067] By normalizing the wave assessment value PZ, the change assessment value BZ, and the trend assessment value QP, and substituting them into the preset formula: JPZ=PZ×d1+BZ×d2+QP×d3, the health assessment value JPZ is obtained, where d1, d2, and d3 are preset weight coefficients;

[0068] For each household appliance, three health assessment value intervals are preset, namely, the health assessment value safety interval, the health assessment value warning interval, and the health assessment value maintenance interval, and the health assessment value corresponding to the detection period is matched with the three health assessment value intervals;

[0069] If the health assessment value matches the safe range, it means that the household appliance is in normal operation and no processing is required;

[0070] If the health assessment value warning interval is matched, it means that the household appliance is in a slightly abnormal state. At this time, a warning instruction containing the name of the household appliance is generated and sent to the user's mobile terminal. After receiving the warning instruction, the user performs maintenance and inspection on the household appliance corresponding to the warning instruction; for example: carefully check whether the placement of the household appliance is stable to ensure that it will not produce additional vibration or displacement due to unstable placement, affecting normal operation; check whether the placement of the household appliance is reasonable to avoid adverse effects on the performance of the appliance due to surrounding environmental factors (such as proximity to heat sources, humid areas, etc.); also pay attention to check whether there is a lot of dust attached to the inside of the heat dissipation port, because dust accumulation may hinder heat dissipation, causing the appliance to overheat, thereby affecting its working efficiency and lifespan; after completing the inspection, take corresponding maintenance measures according to the actual situation, such as adjusting the position of the appliance, cleaning the dust on the heat dissipation port, etc., to ensure that the appliance returns to normal operation;

[0071] If the health assessment value maintenance interval is matched, it means that there is a hidden danger of failure in the household appliance. At this time, the household appliance will be controlled to shut down and stop running to prevent energy loss caused by further deterioration of the fault, and a maintenance instruction containing the name of the household appliance will be generated and sent to the user's mobile terminal. When the user receives the maintenance instruction of the corresponding appliance, he can reply whether to repair it. If the reply is to repair it, a matching maintenance instruction containing the name of the household appliance will be generated and sent to the intelligent matching module;

[0072] The server is used to manage user information and maintenance personnel information;

[0073] User information management includes: storage of user registration information, including name, phone number and home address;

[0074] Maintenance personnel information management: used to register maintenance personnel's maintenance information, including name, phone number, and length of service; at the same time, it establishes a location positioning connection with the maintenance personnel's mobile phone. In addition, it is also responsible for recording the maintenance personnel's maintenance points, of which the maintenance personnel's maintenance points are reset once a month.

[0075] The intelligent matching module calculates the matching value of the maintenance personnel according to the name of the household appliance to be repaired, the maintenance appointment time, the maintenance address, and the address, length of service, and points of the maintenance personnel, and selects the selected maintenance personnel to repair the household appliance according to the matching value of the maintenance personnel. The specific process is as follows:

[0076] When a matching maintenance instruction containing the name of a household appliance is received, the name of the household appliance to be repaired is extracted; the household appliance to be repaired is marked as a device to be repaired, and the user submits an appointment time for the device to be repaired through a mobile terminal;

[0077] Obtain the corresponding maintenance personnel from the server according to the name of the equipment to be repaired, and mark them as preliminary candidates;

[0078] Screen the preliminary candidates who have no maintenance appointment on the appointment date and mark them as candidates Rx, where x=1, ..., p; p is the total number of candidates; the screening process may include: sending information about the appointment for door-to-door maintenance and the maintenance appointment time to the mobile phone of the preliminary candidates, and if the preliminary candidates reply that they accept the maintenance, marking the preliminary candidates as candidates;

[0079] Obtain the user's address and the address of the candidate from the server and calculate the distance difference to obtain the maintenance distance FRx; extract the candidate's maintenance service years SRx and the candidate's maintenance points JR from the server

[0080] After normalizing the maintenance distance FRx, maintenance service years SRx, and maintenance points JRx, the matching value WRx is obtained using the formula: WRx=FRx×e1+SRx×e2+JRx×e3, where e1, e2, and d3 are preset proportional coefficients;

[0081] The candidate with the largest matching value is marked as the selected maintenance personnel of the user, and a message is sent to the selected maintenance personnel's mobile terminal indicating the maintenance appointment time; at the same time, the name and contact number of the selected maintenance personnel are sent to the user's mobile terminal. After the selected maintenance personnel completes the maintenance, a maintenance completion message is sent to the server;

[0082] After receiving the repair completion instruction sent by the selected maintenance personnel, the server generates maintenance score evaluation information and sends it to the user's mobile phone terminal where the selected maintenance personnel last accepted the repair. The score evaluation information includes: whether the door-to-door service is on time (1 to 5 points), the maintenance quality (1 to 5 points) and the service attitude (1 to 5 points);

[0083] After receiving the maintenance score evaluation information, the user scores the maintenance service provided by the selected maintenance personnel, and the total score scored by the user is added to the maintenance score stored in the server by the selected maintenance personnel.

[0084] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. Energy-saving management method based on AI algorithm and multiple collection methods, The method comprises the following steps: Step 1: Match the initial collection frequency according to the power of the household appliance when it is in use, and switch the collection frequency according to the power jump duration and jump times when the household appliance is running. The household appliance collects the health detection parameters of the household appliance itself according to the collection frequency corresponding to each moment; Step 2: Establish a connection with the cloud AI service platform to receive the health detection parameters of the home appliances, analyze the received health detection parameters, obtain the wave evaluation value, change evaluation value and trend evaluation value, and obtain the health evaluation value after comprehensive analysis of the three. Monitor the operating status of the home appliances according to the health evaluation value. If the monitoring result shows that maintenance is needed, generate a matching maintenance instruction containing the name of the home appliance; Step 3: The server is used to manage user information and maintenance personnel information; Step 4: when receiving a matching maintenance instruction including the name of a household appliance, a matching value of a maintenance person is calculated based on the name of the household appliance to be repaired and the corresponding user information and maintenance person information, and a maintenance person is selected to repair the household appliance based on the matching value of the maintenance person; In step 1, the specific process of matching the initial acquisition frequency according to the power of household appliances in use is as follows: By connecting household appliances to smart sockets, the real-time power of household appliances when in use is obtained. Household appliances include refrigerators, televisions, washing machines, air conditioners, electric water heaters, and electric ovens. For each household appliance, multiple power collection intervals are preset, and each power collection interval corresponds to a basic collection frequency. At the same time, the power collection duration is preset, and the average power of the household appliance during the power collection duration is recorded with the start time of the household appliance as the starting time. The average power of the household appliance during the power collection duration is matched with all power collection intervals, and the corresponding basic collection frequency is output as the initial collection frequency when the household appliance is started. In step 1, the collection frequency is switched according to the power jump duration and the number of jumps. The specific process of the household appliance collecting the health detection parameters of the household appliance itself according to the collection frequency corresponding to each moment is as follows: After the startup time of the household appliance is longer than the preset power collection time, if the power value of the household appliance jumps from the original power collection interval to another power collection interval at a certain moment, then this moment is marked as the power jump moment, and a time jump threshold is preset. Starting from the power jump moment, the duration of the power value of the household appliance jumping from the original power collection interval to another power collection interval is detected, which is recorded as the jump duration. If the jump duration is greater than or equal to the preset time jump threshold, the basic collection frequency of the household appliance is updated to the basic collection frequency corresponding to the current power collection interval; If the jump duration is less than the preset time jump threshold, starting from the power jump moment, the number of times the power value of the household appliance jumps across the power collection interval is detected, recorded as the number of jumps, and the jump number threshold is preset. If the jump number is greater than or equal to the preset jump number threshold, the emergency collection frequency is preset, and the original initial collection frequency of the household appliance is increased to the emergency collection frequency; if the jump number is less than the preset jump number threshold, the collection frequency of the household appliance remains unchanged; When the emergency collection frequency is increased, if the power value of the household appliance remains within a certain power collection interval and the duration is greater than or equal to the preset time jump threshold, the collection frequency of the household appliance is gradually restored from the emergency collection frequency to the basic collection frequency corresponding to the power collection interval; The household appliances collect health detection parameters of the household appliances themselves according to the corresponding collection frequency at each moment; the health detection parameters include: power value, surface temperature value, vibration frequency, and noise value.

2. The energy-saving management method based on AI algorithm and multiple collection methods according to claim 1 is characterized in that: In step 2, the received health detection parameters are analyzed to obtain the specific process of wave evaluation: For each household appliance, the operation time of the household appliance is divided into several continuous and non-overlapping detection time periods at fixed intervals; For each detection period, the standard deviation of each health detection parameter in the detection period is calculated to obtain the health fluctuation value, which includes: power fluctuation value, temperature fluctuation value, vibration fluctuation value and noise fluctuation value; It is preset that each health fluctuation value corresponds to a fluctuation threshold. The health fluctuation value is compared with the corresponding fluctuation threshold. If the health fluctuation value is greater than the corresponding fluctuation threshold, the difference between the health fluctuation value and the corresponding fluctuation threshold is calculated and marked as the health wave difference JKi, where i is the number of the health wave difference, i=, 1, 2...n, n≤4; n is the total number of health wave difference numbers, using the formula: , and obtain the wave evaluation value PZ, where ai is the preset weight coefficient, and ai includes: a1, a2, a3, and a4.

3. The energy-saving management method based on AI algorithm and multiple collection methods according to claim 2 is characterized in that: In step 2, the received health detection parameters are analyzed to obtain the specific process of variable evaluation value: For each health detection parameter, let the value of the health detection parameter at different collection points during the detection period be P g , g is the number of the collection point, g = 1, 2...m; m is the total number of collection points, using the formula: , get the change rate value PH of the health detection parameter, where T g For the collection time; Each health detection parameter is preset to have a corresponding change rate threshold. If the change rate of the health detection parameter is greater than the corresponding threshold, the deviation between the change rate of the health detection parameter and the corresponding threshold is calculated and marked as the health deviation value JPs, where s is the number of the health deviation value, s=, 1, 2...e, e≤4, and e is the total number of health deviation value numbers; using the formula: , and obtain the variable evaluation value BZ, where bi is the preset weight coefficient, bi includes: b1, b2, b3, b4.

4. The energy-saving management method based on AI algorithm and multiple collection methods according to claim 3 is characterized in that: In step 2, the received health detection parameters are analyzed to obtain the trend evaluation value. The specific process is as follows: For each health detection parameter in each detection period, collect the historical data of household appliances in the detection period, use the historical data of the health detection parameters in the detection period to fit the linear regression line equation of the health detection parameters, and for each collection point, calculate its corresponding value on the linear regression line equation of the health detection parameters and mark it as the trend value Y g , the actual value P g and trend value Y g Perform the difference and get the real bias value △P g , using the formula: , get the trend deviation value QD; Obtain the trend deviation value QDr corresponding to each health detection parameter within the detection period, where r is the number of the trend deviation value corresponding to each health detection parameter, r=1, 2...h; h is the total number of trend deviation value numbers; use the formula: , and obtain the trend evaluation value QP, where cr is the preset weight coefficient, cr includes: c1, c2, c3, c4.

5. The energy-saving management method based on AI algorithm and multiple collection methods according to claim 4 is characterized in that: In step 2, the specific process of obtaining the health evaluation value and monitoring the operating status of the household appliance according to the health evaluation value is as follows: By normalizing the wave assessment value PZ, the change assessment value BZ, and the trend assessment value QP, and substituting them into the preset formula: JPZ=PZ×d1+BZ×d2+QP×d3, the health assessment value JPZ is obtained, where d1, d2, and d3 are preset weight coefficients; For each household appliance, three health assessment value intervals are preset, namely, the health assessment value safety interval, the health assessment value warning interval, and the health assessment value maintenance interval, and the health assessment value corresponding to the detection period is matched with the three health assessment value intervals; If the health assessment value is matched to the safety range, no processing is required; If the health assessment value warning interval is matched, a warning instruction containing the name of the household appliance is generated and sent to the user's mobile terminal. After receiving the warning instruction, the user performs maintenance and inspection on the household appliance corresponding to the warning instruction; If the health assessment value matches the maintenance range, the household appliances are shut down and stopped, and a maintenance instruction containing the name of the household appliances is generated and sent to the user's mobile terminal. When the user receives the maintenance instruction for the corresponding appliance, he or she can reply whether to perform the maintenance. If the user replies to perform the maintenance, a matching maintenance instruction containing the name of the household appliance is generated.

6. The energy-saving management method based on AI algorithm and multiple collection methods according to claim 1 is characterized in that: The working process of step three is: User information management includes: storage of user registration information, including name, phone number and home address; Maintenance personnel information management: used to register maintenance personnel's maintenance information, including name, phone number, and length of service; at the same time, a location positioning connection is established with the maintenance personnel's mobile phone. In addition, it is also responsible for recording the maintenance personnel's maintenance points, which are reset once a month.

7. The energy-saving management method based on AI algorithm and multiple collection methods according to claim 6 is characterized in that: The working process of step 4 is: When a matching maintenance instruction containing the name of a household appliance is received, the name of the household appliance to be repaired is extracted; the household appliance to be repaired is marked as a device to be repaired, and the user submits an appointment time for the device to be repaired through a mobile terminal; Obtain the corresponding maintenance personnel from the server according to the name of the equipment to be repaired, and mark them as preliminary candidates; Screen the preliminary candidates who do not have a maintenance appointment on the appointment date and mark them as candidates Rx, x=1, ..., p; p is the total number of candidates to be selected; Obtain the user's address and the address of the candidate from the server and calculate the distance difference to obtain the maintenance distance FRx; extract the candidate's maintenance service years SRx and the candidate's maintenance points JR from the server After normalizing the maintenance distance FRx, maintenance service years SRx, and maintenance points JRx, the matching value WRx is obtained using the formula: WRx=FRx×e1+SRx×e2+JRx×e3, where e1, e2, and d3 are preset proportional coefficients; The candidate with the largest matching value is marked as the selected maintenance personnel of the user, and a message about the maintenance and the maintenance appointment time is sent to the selected maintenance personnel's mobile terminal; at the same time, the name and contact number of the selected maintenance personnel are sent to the user's mobile terminal. After the selected maintenance personnel completes the maintenance, the maintenance completion information is sent to the server.

8. The energy-saving management method based on AI algorithm and multiple collection methods according to claim 7 is characterized in that: The server processes the following after receiving the repair completion information: After receiving the repair completion instruction sent by the selected maintenance personnel, the server generates maintenance score evaluation information and sends it to the user's mobile phone terminal where the selected maintenance personnel last accepted the repair. The score evaluation information includes: whether the door-to-door service is on time (1 to 5 points), the maintenance quality (1 to 5 points) and the service attitude (1 to 5 points); After receiving the maintenance score evaluation information, the user scores the maintenance service provided by the selected maintenance personnel, and the total score scored by the user is added to the maintenance score stored in the server by the selected maintenance personnel.

Images