A good idea buried in a bad idea, the "Kaiser solution" and electrochemical biodiesel.

[NNadir]

I'm not a fan of biofuels, particularly in a time of massive fluctuations in the water supply deriving from the extreme global heating domain under which we all live. Given the land requirements for intensive agriculture, and threats to the food supply that will surely arise as extreme global heating rachets up further, there is probably no worse idea than biofuels. Nevertheless, I've given a lot of thought to an approach to what I call the "Kaiser" solution based on a procedure that works like this: Electrochemical industrial processes - in the Kaiser case it was a choice of when to run aluminum plants relying on the electrochemical Hall Heroult process - only when there is excess electricity on the grid, periods during which electricity is cheap.

That is proposed in this paper:

Distributed Economic Nonlinear Model Predictive Control for Flexible Electrified Biodiesel Production─Part I: Sequential Architecture
Mohammad El Wajeh, Marcel Granderath, Alexander Mitsos, and Adel Mhamdi, Industrial & Engineering Chemistry Research 2024 63 (42), 17997-18012

Although biodiesel is a bad idea, the idea of using excess electricity for electrochemical processes is not a bad idea.

Grid demand is roughly sinusoidal with a periodic fluctuation in amplitude, with the global peak for any day generally occurring in the late afternoon and early evening, around the time the sun is going down.

Here, is a typical demand curve from California on September 5, 2024, a day of extreme temperatures in that state, temperatures so high that one might easily die without access to air conditioning:



Note that the peak occurred as described above; if one looks at these curves on any grid anywhere on any day, the shape, albeit not the amplitude is roughly the same.

Here is the supply curve for that day:



Note the dependence on dangerous natural gas; on the hottest day of the year, dangerous fossil fuel waste was being dumped in the early evening to keep the air conditioners running. Note also that there was a "demand event" that day.

So called "renewable energy" fluctuated wildly during that period; around noon the wind industry in particular was useless:



So called "renewable energy's" output had nothing to do with demand.

A better alternative to this unpredictability would be to build enough nuclear plants to continuously produce 48,000 MW of power in a process intensification procedure (perhaps diverting some of the primary energy, heat, to purposes other than electrical generation) and use the electricity for electrochemical processes when electricity was in excess, diverting it to the grid when grid prices are high enough to justify shutting the electrochemical process. Processes of this type include aluminum manufacture, as with Kaiser, and the FFC Cambridge process for the preparation of titanium and in fact, many other metals (both of these processes require heat as primary energy).

Again, I oppose biodiesel, I like this approach described in the process in the paper, even though it appeals to so called "renewable energy" and not to a superior and cleaner approach, nuclear energy.

To wit from the paper's introduction:

Flexible operation of electrified and renewable-powered chemical plants facilitates optimizing energy consumption and capitalizing on the variability of energy prices, thereby contributing to more economically viable and potentially sustainable chemical production. (1,2) However, the operational time scales of some chemical plants align with the frequency of energy supply fluctuations, leading to prolonged periods of transient states during dynamic operations. (3) Thus, to achieve optimal and feasible flexible operations of such plants, it is crucial to incorporate process dynamics and constraints into frequent scheduling decisions, (4) as hypothetically illustrated in Figure 1. Economic nonlinear model predictive control (eNMPC) emerges as an effective approach for integrated decision-making in chemical plants across various time scales. (5) Unlike traditional hierarchical automation tasks in chemical plants, eNMPC maximizes economic performance while directly controlling the process. (6) Employing a receding horizon approach, eNMPC solves online dynamic optimization (DO) problems with an economic objective over a sufficiently extended time horizon, utilizing an open-loop process model while accounting for process constraints. Several studies in the literature explore the methodological (7−11) and application (12−15) aspects of eNMPC. Ellis et al. (5) provides a comprehensive overview of the principles, methodologies, and practical implications of eNMPC, addressing both closed-loop stability and performance for nonlinear systems. Notably, the real-time tractability of eNMPC is particularly critical for its applications in chemical processes, where it encompasses DO problems involving large-scale nonlinear systems. Therefore, implementing real-time tractable eNMPC necessitates efficient strategies, such as fast update schemes, (16,17) model reduction techniques, (18−21) and/or system decomposition with noncentralized control...(12,22)

That is the right idea if described for a poor and unsustainable process.

Have a nice day tomorrow.

[marble falls]

1. Texas gives away night wind generated electricty.