Energy consumption monitoring and predicting method and system

Abstract

The invention relates to the field of data processing, in particular to an energy consumption monitoring and predicting method and system, which introduces an energy consumption efficiency factor related to real-time energy consumption of equipment and integrates the energy consumption efficiency factor into a global pheromone updating mechanism of an ant colony algorithm, so that a scheduling scheme dynamically deviates to equipment with better energy consumption. Meanwhile, in order to deal with emergencies such as emergency order insertion, a disturbance degree model is constructed and is used for evaluating the priority and urgency of the events. When the disturbance degree exceeds a threshold value, the system triggers rescheduling and adjusts heuristic information, emergencies are processed preferentially, finally a perceptible and adaptive closed-loop optimization system is formed, and the energy efficiency and robustness of scheduling are improved. According to the method, energy consumption efficiency factors are introduced to update ant colony algorithm pheromones, heuristic information is dynamically adjusted according to event disturbance degrees, and closed-loop adaptive scheduling of energy consumption and emergencies is achieved.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a method for monitoring and predicting energy consumption. Background Art

[0002] In modern production workshops, production scheduling is the core link to ensure efficient and low-cost operation of the production process. Its goal is to rationally allocate a series of production tasks to available equipment to optimize key indicators such as total production time, equipment utilization or energy consumption.

[0003] However, existing scheduling methods are mostly static, facing two major challenges. First, equipment performance is not static and can change dynamically due to factors such as aging and wear. This can cause deviations between the preset energy consumption model and actual operating conditions, resulting in poor energy control performance for scheduling solutions generated based on outdated data. Second, dynamic events such as urgent order insertions and sudden equipment failures are common during the production process. Traditional scheduling systems lack real-time perception and rapid response mechanisms, making it difficult to effectively address these events, disrupting production plans and resulting in poor robustness.

[0004] To address these issues, intelligent optimization algorithms such as the ant colony algorithm (ACO) are often used for production scheduling. This algorithm simulates the foraging behavior of ants to find the optimal scheduling path within the solution space. However, in dynamic scenarios like these, the standard ACO suffers from significant drawbacks. First, its pheromone update mechanism typically relies solely on fixed cost models (such as total work hours or estimated energy consumption) and fails to perceive the real-time health status of equipment. When a piece of equipment deteriorates and its actual energy consumption increases, the algorithm fails to detect this and may continue to assign tasks based on its lower historical cost, resulting in a discrepancy between the scheduling results and actual energy consumption. Second, the heuristic information used in the algorithm (i.e., the expected degree of task shifting) is often static and immutable once set. This makes the algorithm slow to respond to unexpected events and unable to intelligently adjust task priorities to prioritize tasks of high importance or urgency. Summary of the Invention

[0005] In response to the problems of deviation between the above-mentioned scheduling results and actual energy consumption and slow response when facing emergencies, in the first aspect, the present invention proposes an energy consumption monitoring and prediction method, including: obtaining a production task set and an available equipment set; scheduling the production tasks based on an ant colony algorithm to generate a scheduling plan; collecting the actual power of each task on the corresponding equipment during the execution of the scheduling plan, and obtaining the expected power calculated based on the equipment benchmark data; calculating the equipment energy efficiency factor, which is negatively correlated with the actual power and positively correlated with the expected power; introducing the equipment energy efficiency factor into the global pheromone update mechanism of the ant colony algorithm, so that the pheromone increment is positively correlated with the average value of the energy efficiency factors of the equipment corresponding to all tasks in the path; responding to emergencies in the production process, calculating the disturbance degree that is positively correlated with the event priority and positively correlated with the deadline urgency; when the disturbance degree is greater than a set threshold, triggering rescheduling and adjusting the heuristic information to give priority to the emergency.

[0006] Compared to prior art scheduling methods that employ static energy consumption models and lack the ability to respond to emergencies, the present invention introduces a device energy efficiency factor based on real-time power monitoring and integrates it into the global pheromone update mechanism of the ant colony algorithm, achieving dynamic perception of the device's true energy consumption status. This allows scheduling decisions to consistently favor the most energy-efficient devices, thereby reducing the total energy consumption of actual production. Furthermore, the present invention constructs a disturbance degree model to quantify the urgency of emergencies. When the disturbance degree exceeds a threshold, it automatically triggers rescheduling and adjusts heuristic information, giving the scheduling system the ability to respond to dynamic changes such as emergency orders or equipment failures, significantly improving the robustness and adaptability of production scheduling.

[0007] Furthermore, the method for calculating the equipment energy efficiency factor specifically includes: ; in Representation device In the execution of the task Energy efficiency factor when Representation device In the execution of the task Real-time power collection at the time; Representation device In the execution of the task Expected power when represents the hyperbolic tangent function; Represents the tuning factor.

[0008] This method uses a hyperbolic tangent function to normalize and nonlinearly map the relative deviation between actual and expected power, enabling the energy efficiency factor to respond smoothly and within bounds to power fluctuations. Compared to simple linear proportional calculations, this approach effectively suppresses the excessive impact of extreme power deviations, making the factor more stable and ensuring the robustness of subsequent pheromone updates.

[0009] Furthermore, the global pheromone update mechanism is specifically as follows: ; in Represents ants On the mission Then select the task The pheromone increment left on the path; represents the pheromone intensity constant; Represents ants The total evaluation cost of the constructed scheduling scheme; Indicates that in ants In the path assigned to the device The average energy efficiency factor of all tasks.

[0010] The global pheromone update mechanism of this invention directly links the average energy efficiency factor of a path to the pheromone increment, providing positive incentives for energy optimization. Unlike traditional ant colony algorithms that only consider path length or fixed costs, this invention allows paths containing devices that have recently demonstrated higher energy efficiency to accumulate more pheromones during the iterative optimization of scheduling solutions, thereby guiding the algorithm to converge more quickly to a globally optimal or suboptimal scheduling solution with lower energy consumption.

[0011] Furthermore, the calculation method of the disturbance degree is specifically as follows: ; in Indicates a new event For tasks that are currently being executed or are about to be executed the degree of disturbance caused; Indicates the priority of the new event; Indicates the average priority of all tasks in the current task pool; Indicates the deadline time point required by the new event relative to the current time; Indicates the current task The expected completion time; and The value range is The weight coefficients of , and the sum is 1.

[0012] The proposed method for calculating disturbance levels comprehensively considers two key dimensions: the priority of the emergency and the urgency of the deadline. These factors are balanced using weighting coefficients, enabling a more comprehensive and objective assessment of the impact of new events on existing plans. Compared to existing technologies that rely solely on a single rule (such as first-come, first-served) to determine whether to reschedule, this proposed method provides a quantifiable and more refined decision-making basis, avoiding unnecessary and frequent rescheduling while ensuring timely responses to high-importance and high-urgency events.

[0013] Furthermore, the adjustment heuristic information specifically includes: ; in represents the adjusted heuristic information; represents the original heuristic information; Indicates the disturbance impact factor with a value greater than 0; Indicates a task As an emergency for the previous task The degree of disturbance.

[0014] The heuristic information adjustment method of this invention directly uses the calculated disturbance level to enhance the transfer expectation of the emergency task. This significantly increases the probability that ants will select this emergency as their next task during rescheduling. Compared to the traditional ant colony algorithm, where heuristic information remains fixed, this dynamic adjustment mechanism directly transforms the urgency of the emergency into the algorithm's search preference, prioritizing it during rescheduling and ensuring it is scheduled as quickly as possible, effectively shortening the response time to the emergency.

[0015] Furthermore, it also includes modeling the production task set as a graph model , where the node set Represents all production tasks, edge set Represents the process transfer relationship allowed between tasks.

[0016] Furthermore, obtaining the production task set and the available equipment set includes: obtaining the production task set and the available equipment set from a manufacturing execution system (MES), a supervisory control and data acquisition system (SCADA), and an enterprise resource planning system (ERP).

[0017] Furthermore, when the actual power is equal to the expected power, the energy consumption efficiency factor is equal to 1; when the actual power is greater than the expected power, the energy consumption efficiency factor is less than 1; when the actual power is less than the expected power, the energy consumption efficiency factor is greater than 1.

[0018] Furthermore, the calculation method of the disturbance degree includes a time urgency term; the time urgency term takes a positive value only when the expected completion time of the current task is later than the deadline of the new event, and takes a value of 0 in other cases.

[0019] In a second aspect, the present invention provides an energy consumption monitoring and prediction system, comprising a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the energy consumption monitoring and prediction method of the present invention is implemented.

[0020] The technical effects of the present invention are: This paper proposes a dynamic intelligent scheduling method. Its core innovation lies in two improvements to the traditional ant colony algorithm: first, it introduces an energy efficiency factor linked to the real-time power of the equipment to dynamically update pheromones, ensuring that scheduling continuously approaches the optimal energy consumption solution. Second, it constructs a disturbance degree model that quantifies the urgency of events. When situations such as emergency interruptions occur, it automatically triggers rescheduling and prioritizes them. This creates a closed-loop adaptive scheduling system that combines energy efficiency optimization with dynamic event response capabilities, significantly improving the efficiency and robustness of production scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 is a flow chart schematically illustrating a method for monitoring and predicting energy consumption according to an embodiment of the present invention; Figure 2 Schematically illustrates a heat map of the device energy efficiency factor in the ninth iteration of an embodiment of the present invention; Figure 3 Schematically shows a pheromone matrix heat map after the 9th iteration of an embodiment of the present invention; Figure 4 10 is a heat map schematically showing the energy efficiency factor of the device in the 10th iteration of the embodiment of the present invention; Figure 5 Schematically shows a pheromone matrix heat map after the 10th iteration of an embodiment of the present invention; Figure 6 FIG. 1 is a block diagram schematically illustrating the structure of an energy consumption monitoring and prediction system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] 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 them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work shall fall within the scope of protection of the present invention.

[0023] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0024] Energy consumption monitoring and prediction method embodiment: like Figure 1 As shown, the energy consumption monitoring and prediction method of the present invention includes: S1. Initialization acquisition and modeling of production tasks and equipment status data.

[0025] First, the system obtains the production task set to be scheduled from the Manufacturing Execution System (MES), Equipment Monitoring and Control System (SCADA) and Enterprise Resource Planning System (ERP) and available device sets For each production task , obtain its process ID, planned output, working hours, delivery date and other static information. , obtain benchmark data such as its rated power and energy consumption model under different working conditions.

[0026] In this embodiment, the complex production scheduling problem can be abstracted into a graph model , where the node set Represents all production tasks to be executed, edge set This modeling process lays the foundation for the subsequent path optimization using the graph search-based ant colony algorithm.

[0027] S2. Construct a dynamic pheromone update mechanism based on real-time energy consumption deviation.

[0028] When updating pheromones, the traditional ant colony algorithm only considers the total length of the paths built by the ants (i.e., total energy consumption or total working hours). This invention introduces the consideration of the deviation between the actual energy consumption of the equipment and the expected energy consumption.

[0029] First, during the production execution process, the system collects real-time data from each device through the SCADA interface. In the execution of the task The actual power At the same time, according to the equipment baseline energy consumption model obtained in step S1, the task In this device Theoretical expected power .

[0030] In order to quantify the energy consumption performance of a device when performing a specific task, the device energy efficiency factor is constructed in this embodiment. This factor is designed to reflect the degree to which the actual energy consumption of a device deviates from the theoretical model. This factor is constructed by comparing the relative deviation between actual power and expected power and using a nonlinear function for smoothing and normalizing, thereby obtaining a stable indicator for measuring efficiency. The specific calculation formula is as follows: ; in Representation device In the execution of the task Energy efficiency factor when Representation device In the execution of the task Real-time power collection at the time; Indicates the device's In the execution of the task Expected power when Represents the hyperbolic tangent function, whose range is (-1,1), which can smooth and normalize the power deviation; It represents the parameter adjustment factor. In this embodiment, the empirical value 1e-6 can be taken to avoid the situation where the denominator is 0.

[0031] From the above formula, we can see that when the actual power Equal to the expected power When the deviation is 0, , This indicates that the energy consumption performance of the equipment is in line with theoretical expectations and the efficiency status is standard. Greater than expected power For example, if the equipment is aging or failing, resulting in increased energy consumption, the power deviation is positive. The value is positive, so will be less than 1, and the larger the deviation, The closer it is to 0, the lower the energy efficiency of the device.

[0032] On the contrary, when the actual power Less than expected power , for example, energy consumption is reduced through technical transformation or optimized operation, and the deviation is negative, The value is negative, will be greater than 1, indicating that the energy efficiency of the equipment is better than expected.

[0033] Furthermore, the equipment energy efficiency factor Integrate it into the global pheromone update rule of the ant colony algorithm. The basis for integration is: when evaluating the quality of a scheduling path, not only its total cost should be considered, but also those paths that use devices with better current status (i.e., higher energy efficiency) should be rewarded. After completing a path construction, that is, generating a complete scheduling sequence, the edges passed by the path The pheromone increment left on It will be corrected by the following formula: ; in Represents ants On the mission Then select the task The pheromone increment left on the path; represents the pheromone intensity constant; Represents ants The total evaluation cost of the constructed scheduling solution, which may be the total energy consumption in one embodiment; Indicates that in ants In the path assigned to the device The average energy efficiency factor of all tasks.

[0034] This pheromone update rule ensures that even if the total cost of the two scheduling solutions is exactly the same, but if a scheme uses more recent energy efficiency (i.e. If a device with a larger value is selected, its corresponding path will receive more pheromone rewards. This will positively guide subsequent ant populations to prefer devices that are currently operating in a more economical and efficient manner, allowing the scheduling plan to dynamically and adaptively favor the use of efficient devices.

[0035] like Figure 2 As shown in the figure, this heat map intuitively shows the real-time energy efficiency factor of each device (M00-M03) when performing different tasks (T00-T14) in the 9th iteration. According to the legend, the red area represents , indicating that the actual power consumption of the device on this task is lower than expected, and the energy efficiency is better. The blue area represents , indicating that the equipment is consuming more energy than expected and may be aging or operating inefficiently.

[0036] like Figure 3 As shown in the figure, this pheromone matrix heat map shows the pheromone concentration distribution on the transfer path between tasks after the 9th iteration. The bright spots in the figure, that is, the yellow and white areas, represent the paths that ants tend to choose. According to the above pheromone update rules , those in Figure 2 Devices that are highly efficient (red) will receive more pheromone rewards. Therefore, this pheromone graph can guide subsequent searches to use the most efficient devices.

[0037] S3. Design a dynamic heuristic information adjustment strategy based on production event disturbances.

[0038] The heuristic information in traditional ant colony algorithms, namely the expected degree of task shifting, is typically based on static data, such as the inverse of task processing time or energy consumption, and cannot quickly respond to unexpected events in the production process. This embodiment introduces the ability to perceive unexpected events in the production process and dynamically adjusts the heuristic information to achieve a fast and intelligent response to the scheduling plan.

[0039] When an emergency occurs during the production process, such as an urgent order insertion, equipment failure, or material delay, the system needs to quantify the degree of disruption to the current scheduling plan. ; in Indicates a new event For tasks that are currently being executed or are about to be executed the degree of disturbance caused; Indicates a new event, such as an urgently inserted task; The task being processed or the next task to be processed in the current scheduling plan; Indicates the priority of a new event, which is assigned by the system based on business rules. For example, urgent orders from VIP customers have a higher priority. Indicates the average priority of all tasks in the current task pool; Indicates the deadline time point required by the new event relative to the current time; Indicates the current task The expected completion time; and denote the weight coefficients of priority and urgency respectively, and .

[0040] The above formula evaluates the disturbance from two dimensions: importance and time urgency. The first term It reflects the relative importance of new events. The higher the priority of the event, the greater the disturbance. The second Reflects the time urgency of the new event. If the completion time of the current task is It is already past the deadline for new events , then the term is positive, indicating that the time conflict is serious and the disturbance is high; conversely, if there is ample time, the term is 0, and no urgency disturbance occurs. The larger it is, the greater the impact of the new event on the existing plan, and the more drastic adjustments the algorithm needs to make.

[0041] When a disturbance When the preset threshold is exceeded, the system triggers a rescheduling. In the new ant colony search process, the heuristic information (Represents from the task Transfer to Task The expected degree of ; in represents the adjusted heuristic information; Represents the original heuristic information, usually the task The inverse of processing energy consumption ; Represents the disturbance impact factor, a positive coefficient used to control the impact intensity of the disturbance degree; Indicates a task Relative to the task The event disturbance degree. Here specifically refers to if the task It is an urgent insertion task, then The calculation will be based on Attributes.

[0042] According to the state transition probability formula of the ant colony algorithm , the above adjustment means that when a high-disturbance urgent task When it appears, all directed tasks Heuristic information about the path will be significantly amplified. This makes the task The probability of being selected will be greatly increased. In this way, the algorithm can respond quickly and prioritize urgent tasks to be inserted into the scheduling sequence, achieving adaptive adjustment to external events.

[0043] like Figure 4 As shown in the figure, an emergency occurs after the scheduling system runs to the 9th iteration. A new urgent task 'T15_URGENT' appears in the task list. The appearance of this urgent task will trigger the event disturbance degree in S3. Since this is a high-priority urgent task, its disturbance is likely to exceed the threshold, thus triggering a rescheduling. During the rescheduling process, the system will calculate the time according to the above formula. Dynamically increase the heuristic information for all paths leading to task 'T15_URGENT'.

[0044] like Figure 5 As shown in the figure, comparing the pheromone heatmaps for the 9th and 10th runs, we can see a significant change in the pheromone distribution. New bright spots and paths appear in the 10th run. This is the result of the ant colony, guided by the adjusted heuristic information, finding the optimal insertion location for the urgent task 'T15_URGENT'.

[0045] S4. Closed-loop dynamic scheduling of execution and feedback.

[0046] Integrate the above steps S1 to S3 to form a dynamic optimization closed-loop system of "planning-execution-perception-adjustment".

[0047] First, before production begins, the system runs an optimization based on the standard ant colony algorithm to generate an initial low-energy consumption scheduling plan and starts production according to this plan; then, the system enters a continuous closed-loop monitoring and adjustment phase. On the one hand, the system continuously executes the logic of step S2 to monitor the energy efficiency factor of each device in real time. , and fine-tune the global pheromone in each iteration based on the results This enables the pheromone environment to continuously and accurately reflect the actual energy consumption status of the equipment; further, the system executes the logic of step S3 in parallel to monitor external production events in real time. Once a disturbance is detected If an event exceeds the threshold, the current schedule will be suspended immediately and a rescheduling will be triggered. During the rescheduling process, the ant colony algorithm will use the Real-time corrected pheromones and quilt Dynamically adjusted heuristic information A new round of path search is carried out; finally, after generating a new scheduling plan that is more suitable for the current situation, the system returns to the stage of continuous execution, monitoring and response, forming a complete dynamic optimization closed loop.

[0048] In summary, this invention, through the aforementioned approach, transforms the traditional static ant colony algorithm into a dynamic, intelligent scheduling system capable of sensing and adapting to changes in the real-world production environment. This system not only dynamically optimizes scheduling paths based on the equipment's real-time energy consumption, but also rapidly and appropriately responds to unexpected events during the production process, thereby achieving continuous and effective energy optimization and robust production execution in complex and ever-changing industrial scenarios.

[0049] Energy consumption monitoring and prediction system embodiment: On the other hand, the present invention also provides an energy consumption monitoring and prediction system. Figure 6 As shown, the energy consumption monitoring and prediction system includes a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, an energy consumption monitoring and prediction method according to the first aspect of the present invention is implemented.

[0050] The energy consumption monitoring and prediction system also includes other components well known to those skilled in the art, such as a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0051] In the present invention, the aforementioned memory can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, the computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store the required information and can be accessed by an application, module, or both. Any such computer storage medium can be part of, accessible to, or connectable to a device. Any application or module described in the present invention can be implemented using computer-readable / executable instructions stored or otherwise maintained by such a computer-readable medium.

[0052] In the description of this specification, "multiple" and "several" mean at least two, such as two, three or more, unless otherwise clearly defined.

[0053] While several embodiments of the present invention have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Numerous modifications, variations, and alternatives will occur to those skilled in the art without departing from the concept and spirit of the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be employed in practicing the present invention.

Claims

1. A method for monitoring and predicting energy consumption, characterized in that: The method comprises: Obtaining a set of production tasks and a set of available equipment; scheduling the production tasks based on an ant colony algorithm to generate a scheduling plan; Collecting the actual power of each task on the corresponding device during the execution of the scheduling plan and obtaining the expected power calculated based on the device benchmark data; calculating the device energy efficiency factor, where the device energy efficiency factor is negatively correlated with the actual power and positively correlated with the expected power; The device energy efficiency factor is introduced into the global pheromone update mechanism of the ant colony algorithm, so that the pheromone increment is positively correlated with the average value of the energy efficiency factors of the devices corresponding to all tasks in the path; In response to emergencies in the production process, a disturbance degree is calculated, which is positively correlated with the event priority and the deadline urgency. When the disturbance degree exceeds a set threshold, rescheduling is triggered and heuristic information is adjusted to prioritize the emergency.

2. The energy consumption monitoring and prediction method according to claim 1, characterized in that: The method for calculating the equipment energy efficiency factor specifically includes: ； in Representation device In the execution of the task Energy efficiency factor when Representation device In the execution of the task Real-time power collection at the time; Representation device In the execution of the task Expected power when represents the hyperbolic tangent function; Represents the tuning factor.

3. The energy consumption monitoring and prediction method according to claim 1, characterized in that: The global pheromone update mechanism is specifically as follows: ； in Represents ants On the mission Then select the task The pheromone increment left on the path; represents the pheromone intensity constant; Represents ants The total evaluation cost of the constructed scheduling scheme; Indicates that in ants In the path assigned to the device The average energy efficiency factor of all tasks.

4. The energy consumption monitoring and prediction method according to claim 1, characterized in that: The calculation method of the disturbance degree is specifically as follows: ； in Indicates a new event For tasks that are currently being executed or are about to be executed the degree of disturbance caused; Indicates the priority of the new event; Indicates the average priority of all tasks in the current task pool; Indicates the deadline time point required by the new event relative to the current time; Indicates the current task The expected completion time; and The value range is The weight coefficients of , and the sum is 1.

5. The energy consumption monitoring and prediction method according to claim 1, characterized in that: The adjustment heuristic information specifically includes: ； in represents the adjusted heuristic information; represents the original heuristic information; Indicates the disturbance impact factor with a value greater than 0; Indicates a task As an emergency for the previous task The degree of disturbance.

6. The energy consumption monitoring and prediction method according to claim 1, characterized in that: The production task set is also modeled as a graph model , where the node set Represents all production tasks, edge set Represents the process transfer relationship allowed between tasks.

7. The energy consumption monitoring and prediction method according to claim 1, characterized in that: Acquiring the production task set and the available equipment set includes: The production task set and available equipment set are obtained from a manufacturing execution system (MES), a supervisory control and data acquisition (SCADA) system, and an enterprise resource planning system (ERP).

8. The energy consumption monitoring and prediction method according to claim 2, characterized in that: When the actual power is equal to the expected power, the energy consumption efficiency factor is equal to 1; when the actual power is greater than the expected power, the energy consumption efficiency factor is less than 1; when the actual power is less than the expected power, the energy consumption efficiency factor is greater than 1.

9. The energy consumption monitoring and prediction method according to claim 4, characterized in that: The calculation method of the disturbance degree includes a time urgency term; the time urgency term takes a positive value only when the expected completion time of the current task is later than the deadline of the new event, and takes a value of 0 in other cases.

10. An energy consumption monitoring and prediction system, characterized in that: The method comprises a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the energy consumption monitoring and prediction method according to any one of claims 1 to 9 is implemented.

Images