Energy consumption data monitoring and early warning system and method based on big data

Abstract

The invention discloses an energy consumption data monitoring and early warning system and method based on big data, and relates to the technical field of big data analysis, and the method comprises the steps: collecting operation parameters of abnormal parts in historical maintenance records, calculating an operation score, and determining an operation score threshold value of a fault type; collecting maintenance duration and maintenance space parameters of historical maintenance records, calculating a maintenance score, and establishing a maintenance score prediction formula; setting a training period, calculating the average visitor flow rate of each time period, setting a plurality of sampling time points, calculating operation scores of the sampling time points, setting characteristic duration, calculating change values of the operation scores, and establishing a degradation formula of a certain part; real-time numerical values of operation parameters of a certain part are collected, and the characteristic duration of each fault type is predicted; and calculating maintenance loss scores of the time periods, selecting the time period with the lowest maintenance loss score, giving out early warning, and formulating an optimal early warning strategy.

Description

Technical Field

[0001] The present invention relates to the technical field of big data analysis, and in particular to an energy consumption data monitoring and early warning system and method based on big data. Background Art

[0002] With the development of business intelligence and the increasing demand for energy management, various retail, catering, and service industries are increasingly demanding refined energy consumption monitoring and early warning management. However, traditional energy consumption management methods mainly rely on fixed threshold strategies. This means setting energy consumption limits for each device or area based on historical experience. Once the monitored value exceeds the threshold range, an alarm is triggered. In practical applications, this method has problems such as poor adaptability, high false alarm rate, and inability to accurately identify abnormal energy consumption. It is difficult to meet the refined management needs of modern business scenarios.

[0003] With the development of big data and artificial intelligence technologies, data-driven energy consumption monitoring and early warning mechanisms have emerged. These can not only dynamically adjust energy consumption thresholds and accurately identify abnormal energy consumption, but also effectively reduce energy waste and improve energy efficiency. However, current practices often involve direct shutdown or manual intervention upon detecting energy consumption anomalies. This can lead to inappropriate maintenance timing, disrupt normal operations, and even increase energy loss.

[0004] Therefore, how to conduct intelligent analysis based on energy consumption data, accurately predict abnormal energy consumption trends, formulate optimal early warning strategies, optimize maintenance plans while ensuring normal operations, and improve the intelligence level of energy management has become a problem that needs to be solved at present. Summary of the Invention

[0005] The purpose of the present invention is to provide an energy consumption data monitoring and early warning system and method based on big data to solve the problems raised in the prior art.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for monitoring and early warning of energy consumption data based on big data, the method comprising:

[0007] Step S100: Obtain the operating equipment in a certain business location, determine the operating equipment corresponding to a certain core equipment, determine the equipment combination, monitor the operating status of the equipment combination, and when the operating status is abnormal, arrange staff to perform maintenance and generate a maintenance record;

[0008] Step S200: Obtaining the fault type of the historical maintenance record, summarizing the historical maintenance records of a certain fault type, collecting the operating parameters of the abnormal parts in the historical maintenance record, calculating the operating score of the historical maintenance record, and determining the operating score threshold of the fault type;

[0009] Step S300: collecting maintenance duration and maintenance space parameters of historical maintenance records, calculating maintenance scores for the historical maintenance records, using the maintenance scores and operation scores as fault data groups, summarizing the fault data groups of a certain fault type, and performing function fitting to establish a maintenance score prediction formula for the certain fault type;

[0010] Step S400: Setting a training cycle, dividing a day into several time periods, calculating the average passenger flow in each time period, setting several sampling time points, collecting the operating parameters of a certain part at a certain sampling time point, calculating the operating score at the sampling time point, setting a characteristic duration, calculating the change value of the part operating score during a certain characteristic duration, using the change value and the characteristic duration as a degradation data set, summarizing the degradation data set of the part, performing function fitting, and establishing a degradation formula for the part;

[0011] Step S500: collecting real-time values of operating parameters of a component, calculating a real-time operating score, predicting the characteristic duration of each type of fault occurring in the component, and determining the real-time time points at which all types of faults occur;

[0012] Step S600: Calculate the maintenance loss score of each time period according to the time period between the real-time time points of the fault type, select the time period with the lowest maintenance loss score, and issue an early warning.

[0013] Furthermore, step S100 includes:

[0014] Step S101: Obtain operating equipment in a certain business location, collect core equipment of a certain operating equipment, and number the core equipment;

[0015] Step S102: Summarize the running devices corresponding to the core devices with the same number to obtain the device combination of a core device;

[0016] Step S103: When an abnormality occurs in the equipment combination, the staff performs maintenance and generates a maintenance record, which is uploaded to the maintenance management system;

[0017] By numbering and classifying the core components of running equipment, equipment combinations with the same core components can be quickly identified, allowing managers to track and maintain equipment more efficiently;

[0018] The automatic generation and uploading of maintenance records makes the maintenance process transparent, facilitates subsequent tracking and analysis, and helps optimize equipment maintenance strategies and reduce equipment failure rates.

[0019] Furthermore, step S200 includes:

[0020] Step S201: Obtain historical maintenance records of a certain equipment combination through the maintenance management system, collect the fault type of a certain historical maintenance record, and summarize historical maintenance records of the same fault type;

[0021] Step S202: Collect the operating parameters of the abnormal parts in a certain historical maintenance record, and calculate the operating score of the historical maintenance record according to the following operating score formula:

[0022]

[0023] Among them, B represents the operation score of historical maintenance records, D b Expressed as the value of the bth operating parameter, d b It is represented as the weight of the bth operating parameter, and C is represented as the total number of operating parameters;

[0024] Step S203: Summarize the operation scores of all historical maintenance records of the same fault type, calculate the average value of the operation score as Z and the standard deviation as V, and calculate the operation score threshold of the fault type as Q=Zk×V, where Q represents the operation score threshold and k represents the preset value;

[0025] By summarizing the historical maintenance records of the same fault type and calculating the mean and standard deviation of the operation score, the operation score threshold can be determined more accurately;

[0026] Utilizing historical data in the maintenance management system and combining it with mathematical statistical methods for analysis makes equipment maintenance more intelligent and helps establish more complete predictive maintenance.

[0027] Furthermore, step S300 includes:

[0028] Step S301: In a set of historical maintenance records of the same fault type, the maintenance duration and maintenance space parameters of a certain historical maintenance record are collected, and the maintenance score of the historical maintenance record is calculated according to the following formula:

[0029]

[0030] Among them, A represents the maintenance score of the historical maintenance record, T represents the maintenance time, m represents the area occupied by the maintenance, M represents the total area, and a T Expressed as the weight of maintenance time, a m Expressed as the weight of the maintenance space;

[0031] Step S302: Collect the maintenance scores and operation scores of all historical maintenance records, use the maintenance scores and operation scores as a fault data group, summarize the fault data group of a certain fault type, and perform function fitting to establish a maintenance score prediction formula for a certain fault type:

[0032] A=B×e+f

[0033] Among them, e represents the weight of maintenance score prediction, and f represents the deviation value of maintenance score prediction;

[0034] The maintenance score is calculated by the maintenance time and maintenance space parameters, which can quantify the maintenance complexity of different fault types.

[0035] Combining statistical analysis of maintenance scores and operation scores makes maintenance management more data-driven, avoids relying solely on experience for maintenance decisions, and improves the accuracy and rationality of maintenance work.

[0036] Furthermore, step S400 includes:

[0037] Step S401: Select a number of consecutive days as a training cycle, divide a day into a number of time periods, and deploy a number of sensors in a business premises to monitor the flow of people in the business premises;

[0038] Step S402: Collect the flow of people in a certain time period on a certain day, summarize the flow of people in a certain time period during the training cycle, and calculate the average flow of people in the time period according to the following formula:

[0039]

[0040] Among them, N represents the average flow of people in the time period, and the Nth n It is represented by the flow of people in the time period corresponding to the nth day, and E is represented by the number of days in the training cycle;

[0041] Step S403: During the training cycle, several sampling time points are set. At a certain sampling time point, the operating parameters of a certain part in a certain equipment combination are collected. The operating parameters are input into an operation scoring formula to calculate the operation score at the sampling time point. When an abnormality occurs in a certain part, the sampling time point is set as the abnormality time point of the part.

[0042] Step S404: Count the abnormal time points of a certain part, set the time interval between every two adjacent abnormal time points as the characteristic duration, calculate the change value of the part operation score during the characteristic duration, use the change value and the characteristic duration as a degradation data group, summarize the degradation data group of the part, perform function fitting, and establish a degradation formula for the part as follows: F = P × G + H, where P represents the characteristic duration, F represents the change value of the part operation score, G represents the weight of the degradation formula, and H represents the deviation value of the degradation formula;

[0043] By monitoring the flow of people in business premises and calculating the average flow of people in each time period, maintenance can be scheduled during low-flow periods to reduce the impact on business operations and improve the rationality of maintenance work;

[0044] By counting the abnormal time points of parts, calculating the change value of the running score, and fitting the function based on the characteristic duration, a degradation formula is established to predict the degradation trend of parts in advance;

[0045] Using sensors to collect equipment operating data and predicting degradation through statistical analysis and mathematical modeling, maintenance shifts from reactive repairs to proactive preventive and predictive maintenance, reducing unplanned downtime and improving equipment availability.

[0046] By monitoring and analyzing the operating conditions of parts, maintenance can be performed at the most appropriate time, avoiding premature replacement of still usable parts and preventing excessive deterioration of parts from causing chain failures, reducing maintenance costs and extending the overall life of the equipment.

[0047] Furthermore, step S500 includes:

[0048] Step S501: Collect the real-time value of the operating parameter of a component of a certain equipment combination, input the real-time value of the operating parameter into the operation scoring formula, calculate the real-time operation score of the component, obtain the operation score threshold of a certain fault type, and calculate the real-time change value of the component according to the following formula:

[0049] F1=Q-B1

[0050] Where F1 represents the real-time change value, and B1 represents the real-time operation score. The real-time change values of the parts corresponding to all fault types are summarized, and the fault types corresponding to the real-time change values are screened. The real-time change values of the parts corresponding to the fault types are input into the degradation formula of the parts to predict the characteristic duration of the fault type of the parts.

[0051] Step S502: Obtain the current time point, add the characteristic duration to the current time period, calculate the real-time time point at which the component fault type occurs, summarize the real-time time points at which all fault types occur, and arrange the fault types in chronological order of the real-time time points;

[0052] Utilize real-time data to calculate the operational scores of parts, and screen parts that may fail by calculating real-time change values, thereby more accurately identifying potential failures and improving prediction accuracy.

[0053] By inputting real-time change values into the degradation formula, the characteristic duration of the part can be calculated, and the specific time point when the part will fail can be predicted, thus achieving proactive management of the equipment health status;

[0054] By predicting the occurrence time of all fault types and arranging them in chronological order, operation and maintenance personnel can rationally arrange maintenance plans, prioritize emergency faults, avoid resource waste, and improve maintenance efficiency.

[0055] This method schedules maintenance in advance through fault prediction, reduces equipment downtime caused by sudden failures, and improves overall equipment availability and production efficiency.

[0056] Furthermore, step S600 includes:

[0057] Step S601: Extract the time period from the current time point to the real-time time point of the first fault type, obtain the average passenger flow in the time period, and calculate the maintenance loss score of the time period according to the following formula:

[0058] X=N×i+A'×r

[0059] Where X represents the maintenance loss score, i represents the weight of the average traffic flow, A' represents the maintenance score corresponding to the fault type, and r represents the weight of the maintenance score;

[0060] Step S602: Extract the time period where two adjacent fault types exist in real time, calculate the maintenance loss score of each time period, summarize the maintenance loss scores of all time periods, select the time period with the lowest maintenance loss score, issue an early warning, and prompt staff to perform maintenance;

[0061] By combining traffic flow monitoring and fault prediction, the system calculates the maintenance loss score for each time period and selects the time period with the lowest loss score for maintenance, thereby minimizing the impact on business operations and ensuring that maintenance work is efficient and does not interfere with normal production or services.

[0062] Avoiding maintenance during periods of high traffic or production loads in business premises can help reduce customer wait times or the risk of production interruptions, improving customer experience and business continuity.

[0063] By analyzing the predicted occurrence time of different faults, the optimal maintenance time period is calculated, and early warnings are issued at the appropriate time to remind maintenance personnel to carry out repairs, thus avoiding emergency repairs and improving the planning and scientific nature of maintenance work;

[0064] The maintenance loss score combines traffic flow factors with maintenance scores to make maintenance decisions more data-driven, improve resource utilization, and select maintenance time periods with low business impact. This can avoid sudden equipment failures during high load periods, reduce unplanned downtime, and improve overall equipment availability and production efficiency.

[0065] Continuously collect pedestrian flow data and equipment operation data, continuously optimize maintenance strategies, make maintenance scheduling more and more accurate, and provide data support for long-term intelligent maintenance systems.

[0066] In order to better implement the above method, a big data-based energy consumption data monitoring and early warning system is proposed. The system includes a maintenance management module, an operation score threshold module, a maintenance score prediction formula module, a degradation formula module, a real-time time point module, and a time period selection module.

[0067] Maintenance management module: obtains the operating equipment in a certain business location, determines the operating equipment corresponding to a certain core equipment, determines the equipment combination, monitors the operating status of the equipment combination, and arranges staff to perform maintenance when the operating status is abnormal and generates maintenance records;

[0068] Operation score threshold module: obtains the fault type of historical maintenance records, summarizes the historical maintenance records of a certain fault type, collects the operating parameters of abnormal parts in the historical maintenance records, calculates the operation score of the historical maintenance records, and determines the operation score threshold of the fault type;

[0069] Maintenance score prediction formula module: collects maintenance time and maintenance space parameters from historical maintenance records, calculates maintenance scores for the historical maintenance records, uses the maintenance scores and operation scores as fault data groups, aggregates fault data groups of a certain fault type, and performs function fitting to establish a maintenance score prediction formula for a certain fault type;

[0070] Degradation formula module: Set a training cycle, divide a day into several time periods, calculate the average traffic flow in each time period, set several sampling time points, collect the operating parameters of a part at a certain sampling time point, calculate the operating score at the sampling time point, set a characteristic duration, calculate the change value of the part's operating score during a certain characteristic duration, use the change value and characteristic duration as a degradation data group, summarize the degradation data group of the part, perform function fitting, and establish a degradation formula for a certain part;

[0071] Real-time time point module: collects the real-time values of a part's operating parameters, calculates the real-time operating score, predicts the characteristic duration of each fault type of the part, and determines the real-time time point when all fault types occur;

[0072] Time period selection module: Calculate the maintenance loss score of each time period based on the time period between the real-time time points of the fault type, select the time period with the lowest maintenance loss score, and issue an early warning.

[0073] Furthermore, the degradation formula module includes a unit for calculating the average flow of people and a unit for establishing a degradation formula:

[0074] Calculate average traffic flow unit: select a number of consecutive days as a training cycle, divide a day into several time periods, place several sensors in the business premises to monitor the traffic flow of the business premises, collect the traffic flow in a certain time period on a certain day, summarize the traffic flow in a certain time period during the training cycle, and calculate the average traffic flow in the time period;

[0075] Establish a degradation formula unit: During the training cycle, set several sampling time points. At a certain sampling time point, collect the operating parameters of a part in a certain equipment combination, input the operating parameters into the operating score formula, and calculate the operating score of the sampling time point. When a part is abnormal, set the sampling time point as the abnormal time point of the part, count the abnormal time points of a part, set the time interval between each two adjacent abnormal time points as the characteristic duration, calculate the change value of the part's operating score in a certain characteristic duration, use the change value and the characteristic duration as the degradation data group, summarize the degradation data group of the part, perform function fitting, and establish a degradation formula for a part.

[0076] Furthermore, the real-time point module includes a feature duration prediction unit and a real-time point determination unit:

[0077] Feature duration prediction unit: collects the real-time value of an operating parameter of a part of a certain equipment combination, inputs the real-time value of the operating parameter into the operating score formula, calculates the real-time operating score of the part, obtains the operating score threshold of a certain fault type, calculates the real-time change value of the part, summarizes the real-time change values of the parts corresponding to all fault types, filters the fault types corresponding to positive real-time change values, inputs the real-time change values of the parts corresponding to the fault types into the degradation formula of the parts, and predicts the feature duration of the part when the fault type occurs;

[0078] Determine the real-time time point unit: obtain the current time point, add the characteristic duration to the current time period, calculate the real-time time point of the fault type of the part, summarize the real-time time points of all fault types, and arrange the fault types in chronological order of the real-time time points.

[0079] Compared with the prior art, the present invention has the following beneficial effects:

[0080] Crowd monitoring is used to calculate the average traffic volume in different time periods. This is combined with equipment failure prediction to rationally schedule maintenance times. The impact of maintenance on business is calculated using maintenance loss scores, allowing for optimal maintenance timing and avoiding periods of high business impact.

[0081] By analyzing fault data groups and establishing a predictive model, maintenance can be shifted from passive repair to active prevention, improving the scientific nature of operation and maintenance. Real-time monitoring combined with mathematical modeling can be used to identify high-energy consumption anomalies in advance, reduce additional energy consumption caused by equipment failures, and optimize equipment operating conditions to improve overall energy utilization and achieve energy conservation and consumption reduction goals. Through the early warning system, staff can be reminded in advance when abnormal trends in equipment occur, improving the scientific nature and proactive nature of maintenance work. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 Schematic diagram of the process of the energy consumption data monitoring and early warning method based on big data of the present invention;

[0083] Figure 2 It is a structural diagram of the energy consumption data monitoring and early warning system based on big data of the present invention. DETAILED DESCRIPTION

[0084] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0085] See also Figure 1 The present invention provides a technical solution: a method for monitoring and early warning of energy consumption data based on big data, the method comprising:

[0086] Step S100: Obtain the operating equipment in a certain business location, determine the operating equipment corresponding to a certain core equipment, determine the equipment combination, monitor the operating status of the equipment combination, and when the operating status is abnormal, arrange staff to perform maintenance and generate a maintenance record;

[0087] Wherein, step S100 includes:

[0088] Step S101: Obtain operating equipment in a certain business location, collect core equipment of a certain operating equipment, and number the core equipment;

[0089] Step S102: Summarize the running devices corresponding to the core devices with the same number to obtain the device combination of a core device;

[0090] Step S103: When an abnormality occurs in the equipment combination, the staff performs maintenance and generates a maintenance record, which is uploaded to the maintenance management system;

[0091] For example, in central air conditioning, the core equipment of the indoor unit is the external unit, and the core equipment of the freezer is the compressor.

[0092] Step S200: Obtaining the fault type of the historical maintenance record, summarizing the historical maintenance records of a certain fault type, collecting the operating parameters of the abnormal parts in the historical maintenance record, calculating the operating score of the historical maintenance record, and determining the operating score threshold of the fault type;

[0093] Wherein, step S200 includes:

[0094] Step S201: Obtain historical maintenance records of a certain equipment combination through the maintenance management system, collect the fault type of a certain historical maintenance record, and summarize historical maintenance records of the same fault type;

[0095] Step S202: Collect the operating parameters of the abnormal parts in a certain historical maintenance record, and calculate the operating score of the historical maintenance record according to the following operating score formula:

[0096]

[0097] Among them, B represents the operation score of historical maintenance records, D b Expressed as the value of the bth operating parameter, d b It is represented as the weight of the bth operating parameter, and C is represented as the total number of operating parameters;

[0098] Step S203: Summarize the operation scores of all historical maintenance records of the same fault type, calculate the average value of the operation score as Z and the standard deviation as V, and calculate the operation score threshold of the fault type as Q=Zk×V, where Q represents the operation score threshold and k represents the preset value;

[0099] For example, if bearing overheating is selected as the fault type, data related to bearing overheating in all historical maintenance records is extracted. The operating parameters of a historical maintenance record are collected: temperature is 85°C, vibration is 3.2 mm / s, and pressure is 5.0 MPa. The weight of temperature is 0.4, the weight of vibration is 0.3, and the weight of pressure is 0.3. The calculated operating score is 36.46.

[0100] Assume that the operation scores obtained from the historical maintenance records of bearing overheating failures are as follows: B1 = 36.46, B2 = 38.2, B3 = 35.8, B4 = 40.1, B5 = 37.3, the calculated average is 37.57 and the standard deviation is 1.5. Assuming k = 1.5, the operation score threshold is 35.32.

[0101] Step S300: collecting maintenance duration and maintenance space parameters of historical maintenance records, calculating maintenance scores for the historical maintenance records, using the maintenance scores and operation scores as fault data groups, summarizing the fault data groups of a certain fault type, and performing function fitting to establish a maintenance score prediction formula for the certain fault type;

[0102] Wherein, step S300 includes:

[0103] Step S301: In a set of historical maintenance records of the same fault type, the maintenance duration and maintenance space parameters of a certain historical maintenance record are collected, and the maintenance score of the historical maintenance record is calculated according to the following formula:

[0104]

[0105] Among them, A represents the maintenance score of the historical maintenance record, T represents the maintenance time, m represents the area occupied by the maintenance, M represents the total area, and a T Expressed as the weight of maintenance time, a m Expressed as the weight of the maintenance space;

[0106] Step S302: Collect the maintenance scores and operation scores of all historical maintenance records, use the maintenance scores and operation scores as a fault data group, summarize the fault data group of a certain fault type, and perform function fitting to establish a maintenance score prediction formula for a certain fault type:

[0107] A=B×e+f

[0108] Among them, e represents the weight of the maintenance score prediction, and f represents the deviation value of the maintenance score prediction.

[0109] Step S400: Setting a training cycle, dividing a day into several time periods, calculating the average passenger flow in each time period, setting several sampling time points, collecting the operating parameters of a certain part at a certain sampling time point, calculating the operating score at the sampling time point, setting a characteristic duration, calculating the change value of the part operating score during a certain characteristic duration, using the change value and the characteristic duration as a degradation data set, summarizing the degradation data set of the part, performing function fitting, and establishing a degradation formula for the part;

[0110] Wherein, step S400 includes:

[0111] Step S401: Select a number of consecutive days as a training cycle, divide a day into a number of time periods, and deploy a number of sensors in a business premises to monitor the flow of people in the business premises;

[0112] Step S402: Collect the flow of people in a certain time period on a certain day, summarize the flow of people in a certain time period during the training cycle, and calculate the average flow of people in the time period according to the following formula:

[0113]

[0114] Among them, N represents the average flow of people in the time period, and the Nth n It is represented by the flow of people in the time period corresponding to the nth day, and E is represented by the number of days in the training cycle;

[0115] Step S403: During the training cycle, several sampling time points are set. At a certain sampling time point, the operating parameters of a certain part in a certain equipment combination are collected. The operating parameters are input into an operation scoring formula to calculate the operation score at the sampling time point. When an abnormality occurs in a certain part, the sampling time point is set as the abnormality time point of the part.

[0116] Step S404: Count the abnormal time points of a certain part, set the time interval between every two adjacent abnormal time points as the characteristic duration, calculate the change value of the part operation score during the characteristic duration, use the change value and the characteristic duration as a degradation data group, summarize the degradation data group of the part, perform function fitting, and establish a degradation formula for the part as follows: F = P × G + H, where P represents the characteristic duration, F represents the change value of the part operation score, G represents the weight of the degradation formula, and H represents the deviation value of the degradation formula;

[0117] For example, suppose

[0118]

[0119]

[0120] The change values are 2.75 and 3.40 respectively. Function fitting is performed and the degradation formula is obtained as F=0.8P+0.5.

[0121] Step S500: collecting real-time values of operating parameters of a component, calculating a real-time operating score, predicting the characteristic duration of each type of fault occurring in the component, and determining the real-time time points at which all types of faults occur;

[0122] Wherein, step S500 includes:

[0123] Step S501: Collect the real-time value of the operating parameter of a component of a certain equipment combination, input the real-time value of the operating parameter into the operation scoring formula, calculate the real-time operation score of the component, obtain the operation score threshold of a certain fault type, and calculate the real-time change value of the component according to the following formula:

[0124] F1=Q-B1;

[0125] Where F1 represents the real-time change value, and B1 represents the real-time operation score. The real-time change values of the parts corresponding to all fault types are summarized, and the fault types corresponding to the real-time change values are screened. The real-time change values of the parts corresponding to the fault types are input into the degradation formula of the parts to predict the characteristic duration of the fault type of the parts.

[0126] Step S502: Obtain the current time point, add the characteristic duration to the current time period, calculate the real-time time point at which the fault type of the part occurs, summarize the real-time time points at which all fault types occur, and arrange the fault types in chronological order according to the real-time time points.

[0127] Step S600: Calculate the maintenance loss score of each time period based on the time period between the real-time time points of the fault type, select the time period with the lowest maintenance loss score, and issue an early warning;

[0128] Step S600 includes:

[0129] Step S601: Extract the time period from the current time point to the real-time time point of the first fault type, obtain the average passenger flow in the time period, and calculate the maintenance loss score of the time period according to the following formula:

[0130] X=N×i+A'×r

[0131] Where X represents the maintenance loss score, i represents the weight of the average traffic flow, A' represents the maintenance score corresponding to the fault type, and r represents the weight of the maintenance score;

[0132] Step S602: Extract the time period where every two adjacent fault types exist in real time, calculate the maintenance loss score of the time period, summarize the maintenance loss scores of all time periods, select the time period with the lowest maintenance loss score, issue an early warning, and prompt the staff to perform maintenance.

[0133] In order to better implement the above method, a big data-based energy consumption data monitoring and early warning system is proposed. The system includes a maintenance management module, an operation score threshold module, a maintenance score prediction formula module, a degradation formula module, a real-time time point module, and a time period selection module.

[0134] Maintenance management module: obtains the operating equipment in a certain business location, determines the operating equipment corresponding to a certain core equipment, determines the equipment combination, monitors the operating status of the equipment combination, and arranges staff to perform maintenance when the operating status is abnormal and generates maintenance records;

[0135] Operation score threshold module: obtains the fault type of historical maintenance records, summarizes the historical maintenance records of a certain fault type, collects the operating parameters of abnormal parts in the historical maintenance records, calculates the operation score of the historical maintenance records, and determines the operation score threshold of the fault type;

[0136] Maintenance score prediction formula module: collects maintenance time and maintenance space parameters from historical maintenance records, calculates maintenance scores for the historical maintenance records, uses the maintenance scores and operation scores as fault data groups, aggregates fault data groups of a certain fault type, and performs function fitting to establish a maintenance score prediction formula for a certain fault type;

[0137] Degradation formula module: Set a training cycle, divide a day into several time periods, calculate the average traffic flow in each time period, set several sampling time points, collect the operating parameters of a part at a certain sampling time point, calculate the operating score at the sampling time point, set a characteristic duration, calculate the change value of the part's operating score during a certain characteristic duration, use the change value and characteristic duration as a degradation data group, summarize the degradation data group of the part, perform function fitting, and establish a degradation formula for a certain part;

[0138] The degradation formula module includes the unit for calculating the average flow of people and the unit for establishing the degradation formula:

[0139] Calculate average traffic flow unit: select a number of consecutive days as a training cycle, divide a day into several time periods, place several sensors in the business premises to monitor the traffic flow of the business premises, collect the traffic flow in a certain time period on a certain day, summarize the traffic flow in a certain time period during the training cycle, and calculate the average traffic flow in the time period;

[0140] Establish a degradation formula unit: During the training cycle, set several sampling time points. At a certain sampling time point, collect the operating parameters of a part in a certain equipment combination, input the operating parameters into the operating score formula, and calculate the operating score of the sampling time point. When a part is abnormal, set the sampling time point as the abnormal time point of the part, count the abnormal time points of a part, set the time interval between each two adjacent abnormal time points as the characteristic duration, calculate the change value of the part's operating score in a certain characteristic duration, use the change value and the characteristic duration as the degradation data group, summarize the degradation data group of the part, perform function fitting, and establish a degradation formula for a part.

[0141] Real-time time point module: collects the real-time values of a part's operating parameters, calculates the real-time operating score, predicts the characteristic duration of each fault type of the part, and determines the real-time time point when all fault types occur;

[0142] The real-time time point module includes a feature duration prediction unit and a real-time time point determination unit:

[0143] Feature duration prediction unit: collects the real-time value of an operating parameter of a part of a certain equipment combination, inputs the real-time value of the operating parameter into the operating score formula, calculates the real-time operating score of the part, obtains the operating score threshold of a certain fault type, calculates the real-time change value of the part, summarizes the real-time change values of the parts corresponding to all fault types, filters the fault types corresponding to positive real-time change values, inputs the real-time change values of the parts corresponding to the fault types into the degradation formula of the parts, and predicts the feature duration of the part when the fault type occurs;

[0144] Determine the real-time time point unit: obtain the current time point, add the characteristic duration to the current time period, calculate the real-time time point of the fault type of the part, summarize the real-time time points of all fault types, and arrange the fault types in chronological order of the real-time time points.

[0145] Time period selection module: Calculate the maintenance loss score of each time period based on the time period between the real-time time points of the fault type, select the time period with the lowest maintenance loss score, and issue an early warning.

[0146] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. The energy consumption data monitoring and early warning method based on big data is characterized by: Methods include: Step S100: Obtain the operating equipment in a certain business location, determine the operating equipment corresponding to a certain core equipment, determine the equipment combination, monitor the operating status of the equipment combination, and when the operating status is abnormal, arrange staff to perform maintenance and generate a maintenance record; Step S200: Obtaining the fault type of the historical maintenance record, summarizing the historical maintenance records of a certain fault type, collecting the operating parameters of the abnormal parts in the historical maintenance record, calculating the operating score of the historical maintenance record, and determining the operating score threshold of the fault type; Step S300: collecting maintenance duration and maintenance space parameters of historical maintenance records, calculating maintenance scores for the historical maintenance records, using the maintenance scores and operation scores as fault data groups, summarizing the fault data groups of a certain fault type, and performing function fitting to establish a maintenance score prediction formula for the certain fault type; Step S400: Setting a training cycle, dividing a day into several time periods, calculating the average passenger flow in each time period, setting several sampling time points, collecting the operating parameters of a certain part at a certain sampling time point, calculating the operating score at the sampling time point, setting a characteristic duration, calculating the change value of the part operating score during a certain characteristic duration, using the change value and the characteristic duration as a degradation data set, summarizing the degradation data set of the part, performing function fitting, and establishing a degradation formula for the part; Step S500: collecting real-time values of operating parameters of a component, calculating a real-time operating score, predicting the characteristic duration of each type of fault occurring in the component, and determining the real-time time points at which all types of faults occur; Step S600: Calculate the maintenance loss score of each time period according to the time period between the real-time time points of the fault type, select the time period with the lowest maintenance loss score, and issue an early warning.

2. The energy consumption data monitoring and early warning method based on big data according to claim 1 is characterized in that: The step S100 includes the following steps: Step S101: Obtain operating equipment in a certain business location, collect core equipment of a certain operating equipment, and number the core equipment; Step S102: Summarize the running devices corresponding to the core devices with the same number to obtain the device combination of a core device; Step S103: When an abnormality occurs in the equipment combination, the staff performs maintenance and generates a maintenance record, which is uploaded to the maintenance management system.

3. The energy consumption data monitoring and early warning method based on big data according to claim 2 is characterized in that: The step S200 includes the following steps: Step S201: Obtain historical maintenance records of a certain equipment combination through the maintenance management system, collect the fault type of a certain historical maintenance record, and summarize historical maintenance records of the same fault type; Step S202: Collect the operating parameters of the abnormal parts in a certain historical maintenance record, and calculate the operating score of the historical maintenance record according to the following operating score formula: Among them, B represents the operation score of historical maintenance records, D b Expressed as the value of the bth operating parameter, d b It is represented as the weight of the bth operating parameter, and C is represented as the total number of operating parameters; Step S203: Summarize the operation scores of all historical maintenance records of the same fault type, calculate the average value Z and the standard deviation V of the operation scores, and calculate the operation score threshold of the fault type as Q=Zk×V, where Q represents the operation score threshold and k represents the preset value.

4. The energy consumption data monitoring and early warning method based on big data according to claim 3 is characterized in that: The step S300 includes the following steps: Step S301: In a set of historical maintenance records of the same fault type, the maintenance duration and maintenance space parameters of a certain historical maintenance record are collected, and the maintenance score of the historical maintenance record is calculated according to the following formula: Among them, A represents the maintenance score of the historical maintenance record, T represents the maintenance time, m represents the area occupied by the maintenance, M represents the total area, and a T Expressed as the weight of maintenance time, a m Expressed as the weight of the maintenance space; Step S302: Collect the maintenance scores and operation scores of all historical maintenance records, use the maintenance scores and operation scores as a fault data group, summarize the fault data group of a certain fault type, and perform function fitting to establish a maintenance score prediction formula for a certain fault type: A=B×e+f Among them, e represents the weight of the maintenance score prediction, and f represents the deviation value of the maintenance score prediction.

5. The energy consumption data monitoring and early warning method based on big data according to claim 4 is characterized in that: The step S400 includes the following steps: Step S401: Select a number of consecutive days as a training cycle, divide a day into a number of time periods, and deploy a number of sensors in a business premises to monitor the flow of people in the business premises; Step S402: Collect the flow of people in a certain time period on a certain day, summarize the flow of people in a certain time period during the training cycle, and calculate the average flow of people in the time period according to the following formula: Among them, N represents the average flow of people in the time period, and the Nth n It is represented by the flow of people in the time period corresponding to the nth day, and E is represented by the number of days in the training cycle; Step S403: During the training cycle, several sampling time points are set. At a certain sampling time point, the operating parameters of a certain part in a certain equipment combination are collected. The operating parameters are input into an operation scoring formula to calculate the operation score at the sampling time point. When an abnormality occurs in a certain part, the sampling time point is set as the abnormality time point of the part. Step S404: Count the abnormal time points of a certain part, set the time interval between every two adjacent abnormal time points as the characteristic duration, calculate the change value of the part operation score in a certain characteristic duration, use the change value and the characteristic duration as the degradation data group, summarize the degradation data group of the part, perform function fitting, and establish the degradation formula of a certain part: F=P×G+H, where P represents the characteristic duration, F represents the change value of the part operation score, G represents the weight of the degradation formula, and H represents the deviation value of the degradation formula.

6. The energy consumption data monitoring and early warning method based on big data according to claim 5 is characterized in that: The step S500 includes the following steps: Step S501: Collect the real-time value of the operating parameter of a component of a certain equipment combination, input the real-time value of the operating parameter into the operation scoring formula, calculate the real-time operation score of the component, obtain the operation score threshold of a certain fault type, and calculate the real-time change value of the component according to the following formula: F1=Q-B1 Where F1 represents the real-time change value, and B1 represents the real-time operation score. The real-time change values of the parts corresponding to all fault types are summarized, and the fault types corresponding to the real-time change values are screened. The real-time change values of the parts corresponding to the fault types are input into the degradation formula of the parts to predict the characteristic duration of the fault type of the parts. Step S502: Obtain the current time point, add the characteristic duration to the current time period, calculate the real-time time point at which the fault type of the part occurs, summarize the real-time time points at which all fault types occur, and arrange the fault types in chronological order according to the real-time time points.

7. The energy consumption data monitoring and early warning method based on big data according to claim 6 is characterized in that: The step S600 includes the following steps: Step S601: Extract the time period from the current time point to the real-time time point of the first fault type, obtain the average passenger flow in the time period, and calculate the maintenance loss score of the time period according to the following formula: X=N×i+A'×r Where X represents the maintenance loss score, i represents the weight of the average traffic flow, A' represents the maintenance score corresponding to the fault type, and r represents the weight of the maintenance score; Step S602: Extract the time period where every two adjacent fault types exist in real time, calculate the maintenance loss score of the time period, summarize the maintenance loss scores of all time periods, select the time period with the lowest maintenance loss score, issue an early warning, and prompt the staff to perform maintenance.

8. An energy consumption data monitoring and early warning system based on big data, used to implement the energy consumption data monitoring and early warning method based on big data according to any one of claims 1 to 7, characterized in that: The system includes a maintenance management module, an operation score threshold module, a maintenance score prediction formula module, a degradation formula module, a real-time time point module, and a time period selection module; The maintenance management module: obtains the operating equipment in a certain business premises, determines the operating equipment corresponding to a certain core equipment, determines the equipment combination, monitors the operating status of the equipment combination, and arranges staff to perform maintenance when the operating status is abnormal and generates maintenance records; The operation score threshold module is configured to obtain the fault type of the historical maintenance records, summarize the historical maintenance records of a certain fault type, collect the operating parameters of the abnormal parts in the historical maintenance records, calculate the operation score of the historical maintenance records, and determine the operation score threshold of the fault type; The maintenance score prediction formula module collects maintenance time and maintenance space parameters from historical maintenance records, calculates maintenance scores for the historical maintenance records, uses the maintenance scores and operation scores as fault data groups, aggregates the fault data groups for a certain fault type, and performs function fitting to establish a maintenance score prediction formula for a certain fault type; The degradation formula module: sets a training cycle, divides a day into several time periods, calculates the average passenger flow in each time period, sets several sampling time points, collects the operating parameters of a part at a certain sampling time point, calculates the operating score at the sampling time point, sets a characteristic duration, calculates the change value of the part operating score during a certain characteristic duration, uses the change value and the characteristic duration as a degradation data group, summarizes the degradation data group of the part, performs function fitting, and establishes a degradation formula for the part; The real-time time point module collects the real-time values of the operating parameters of a certain part, calculates the real-time operating score, predicts the characteristic duration of each fault type of the part, and determines the real-time time point when all fault types occur; The time period selection module calculates the maintenance loss score of the time period according to the time period between the real-time time points of the fault type, selects the time period with the lowest maintenance loss score, and issues an early warning.

9. The energy consumption data monitoring and early warning system based on big data according to claim 8 is characterized in that: The degradation formula module includes a unit for calculating the average flow of people and a unit for establishing a degradation formula: The average pedestrian flow calculation unit: selects a number of consecutive days as a training cycle, divides a day into a number of time periods, arranges a number of sensors in the business premises, monitors the pedestrian flow of the business premises, collects the pedestrian flow in a certain time period of a certain day, summarizes the pedestrian flow in a certain time period during the training cycle, and calculates the average pedestrian flow in the time period; The degradation formula establishment unit is as follows: within a training cycle, several sampling time points are set; at a certain sampling time point, the operating parameters of a certain part in a certain equipment combination are collected; the operating parameters are input into an operating score formula; the operating score of the sampling time point is calculated; when an abnormality occurs in a certain part, the sampling time point is set as the abnormal time point of the part; the abnormal time points of a certain part are counted; the time interval between each two adjacent abnormal time points is set as a characteristic duration; the change value of the operating score of the part in a certain characteristic duration is calculated; the change value and the characteristic duration are used as a degradation data group; the degradation data group of the part is summarized; function fitting is performed; and a degradation formula for a certain part is established.

10. The energy consumption data monitoring and early warning system based on big data according to claim 8 is characterized in that: The real-time time point module includes a feature duration prediction unit and a real-time time point determination unit: The feature duration prediction unit collects the real-time value of an operating parameter of a part of a certain equipment combination, inputs the real-time value of the operating parameter into an operating score formula, calculates the real-time operating score of the part, obtains an operating score threshold of a certain fault type, calculates the real-time change value of the part, summarizes the real-time change values of the parts corresponding to all fault types, selects the fault types corresponding to positive real-time change values, inputs the real-time change value of the part corresponding to the fault type into the degradation formula of the part, and predicts the feature duration of the part when the fault type occurs; The real-time time point determination unit obtains the current time point, adds the characteristic duration to the current time period, calculates the real-time time point of the part failure type, summarizes the real-time time points of all failure types, and arranges the failure types in chronological order according to the real-time time points.

Images