1. Introduction
Elevators, as an indispensable vertical transportation tool in modern high-rise buildings, are closely related to people's lives. Therefore, higher requirements are placed on the static and dynamic characteristics of elevators, such as speed regulation accuracy and speed regulation range. In large-scale high-rise buildings, two or more elevators are usually installed. If the elevators operate independently, it is difficult to improve operating efficiency, which will inevitably lead to a lot of energy waste and also bring great difficulties to the centralized management of elevators.
2. Elevator group control system
Elevator group control system (EGCS) refers to a group of three or more elevators installed in a building as an organic whole, using an automatic control system to schedule the operation of each elevator. The purpose is to improve the operating efficiency of the vertical transportation system, provide services to passengers with shorter waiting time and running time, improve the service quality to passengers, and reduce energy consumption[1].
Elevator group control system is a rather complex system that requires the transmission and processing of hundreds of signals. Currently, there are various ways to implement elevator group control, but the structure of their control systems is similar. Figure 1.1 shows a basic structural block diagram of an elevator group control system. The elevator group control system module receives the external call signal and distributes the call signal to each individual elevator control module according to the processing result of the dispatch strategy algorithm. The individual elevator system module controls the operation of the elevator according to the status of each elevator, the distributed external call signal, and the internal call signal. The elevator group control module is the core of the elevator group control system, responsible for collecting external call signals and coordinating the operation of each elevator [2].
Figure 2.1 Basic structural block diagram of an elevator group control system
3. Simulation of Elevator Group Control System Based on LabVIEW
3.1 LabVIEW software
LabVIEW, short for Laboratory Virtual Instrumentation Engineering Platform, is an innovative software product from National Instruments (NI) in the United States. It is currently the most widely used, fastest-growing, and most powerful graphical software development environment. LabVIEW's key features include: using a graphical programming language to create source code within flowcharts, making it convenient to run and easy to understand.
LabVIEW is a graphical development environment with a compiler that generates optimal code, running at speeds equivalent to pre-written C or C++ programs. LabVIEW's modularity promotes program reusability. It separates interface design from functional design, allowing modifications to the user interface without requiring adjustments to the entire program. LabVIEW uses dataflow diagrams to receive instructions, making programs simple and clear, and fully leveraging the advantages of the G language. This significantly shortens the development cycle of virtual instruments and eliminates the complexities of virtual instrument programming.
3.2 Fuzzy Scheduling
Dispatching is a crucial component of elevator group control systems. The achievement of each control objective of the elevator system relies primarily on the dispatcher's rational allocation of call signals. An elevator group control system is a multi-objective coordinated control system with numerous control indicators, and these objectives are both interrelated and conflicting. For example, reducing elevator system energy consumption increases average waiting time and the rate of long waiting times; lower car crowding leads to longer average waiting times; and shorter average waiting times result in a higher rate of long waiting times.
3.3 Dynamic Partitioning Algorithm
To ensure that each elevator's service area does not overlap and to prevent elevator clustering, an m-story building consisting of n elevators (each with one car) can be divided into n zones. Let the demand on the k-th floor be Uk, and the total demand be U. Each car serves one floor (the main terminal). The zoning is primarily designed for peak-hour/peak-hour patterns, not for random inter-floor patterns. When not in peak-hour/peak-hour patterns, the zoning is automatically canceled, and each elevator car serves the entire building. The specific zoning arrangement is as follows:
Service area of car 1: Floor 1...m1 floor;
Service area of car 2: 1st floor, (m1+1)...m2 floors;
Service area of car 3: 1st floor, (m2+1)...m3 floors;
......
Service area of car j: Floor 1, (mj-1+1)...mj floors;
......
Service area of car n: 1st floor, (mn-1+1)...mn floors;
Where: 1 < m1 < m2 < ... < mj < ... mn ≤ m. The key to dynamic partitioning is to calculate the optimal solutions for n mj (j=1,2,...n). Let the round-trip time of the j-th car be denoted as RTTj, and its calculation formula is as follows:
RTTj=2HjtV+(Sj+1)ts+2Pjtp; j=1, 2,···n
3.4 Elevator Group Control Algorithm Flowchart
The flowchart of the fuzzy group control scheduling algorithm based on dynamic partitioning is shown in Figure 3.1.
Figure 3.1 Flowchart of the dynamic fuzzy optimization scheduling algorithm
When a floor call command is registered, the scheduling algorithm first determines which traffic mode the current system belongs to, selects the values of weight coefficients W1, W2, and W3 under the corresponding traffic mode, calculates the values of each fuzzy variable in response to the call command, calculates each evaluation function value according to the fuzzy decision principle, calculates the total evaluation function value of each elevator in combination with the weight coefficient of the current mode, and finally selects the elevator with the largest total evaluation function value to respond to the call command by comparison, thereby realizing dynamic fuzzy optimization scheduling [3].
3.5 Simulation Experiment
The advantage of this simulation platform is that, apart from the shared hall floor call, the other modules are independent, and the structure of its elevator group control system is shown in Figure 4.7. Therefore, if needed, it can be expanded from 3 elevators to more, only the operating parameters of the newly added elevators need to be initialized; since the scheduling algorithm is an independent sub-VI, it does not affect the expansion of the virtual simulation platform [4].
Figure 3.2 Structural block diagram of elevator group control system
3.6 Experimental System Interface
The simulation environment was LabVIEW 8.6. The interface of the experimental simulation system is shown in Figure 3.3.
Figure 3.3 Experimental simulation interface
3.7 Experimental Results and Analysis
First, 60 random numbers were generated using the LabVIEW "Random Number" function. The first 20 numbers were used to represent outbound calls, the 31st to 40th numbers were for internal selection of elevator A, the 41st to 50th numbers were for internal selection of elevator B, and the 51st to 60th numbers were for internal selection of elevator C, simulating the command registration signals for each floor of the building. Three parameters were used to evaluate the two scheduling algorithms: average waiting time, average riding time, total time taken to respond to all outbound and internal selection signals, and the number of car stops. The experiment used a random inter-floor traffic pattern as an example to verify the superiority of the improved algorithm. The results are as follows:
Figure 3.4 Bar chart of experimental results
The above experimental structure shows that when using fuzzy decision-making and partitioned scheduling in group control scheduling, the average waiting time, average riding time and total number of stops can be effectively reduced, thus improving the efficiency of elevator operation [5].
4. Conclusion
This paper mainly introduces the basic principles of elevator group control systems and uses LabVIEW software to conduct simulation tests on the system. Experimental results show that the group control scheduling optimization algorithm combining dynamic partitioning and fuzzy scheduling can effectively improve the performance of elevators in high-rise buildings.
For more information, please follow the elevator channel.
